Quick viewing(Text Mode)

Combined Estrogen–Progestogen Menopausal Therapy

Combined Estrogen–Progestogen Menopausal Therapy

COMBINED – MENOPAUSAL

Combined estrogen–progestogen menopausal therapy was considered by previous IARC Working Groups in 1998 and 2005 (IARC, 1999, 2007). Since that time, new data have become available, these have been incorporated into the Monograph, and taken into consideration in the present evaluation.

1. Exposure Data 1.1.2 (a) Combined estrogen–progestogen meno- Chem. Abstr. Serv. Reg. No.: 302-22-7 pausal therapy involves the co-administration Chem. Abstr. Name: 17-(Acetyloxy)-6-chlo- of an estrogen and a progestogen to peri- or ropregna-4,6-diene-3,20-dione menopausal women. The use of with IUPAC Systematic Name: 6-Chloro-17-hy- progestogens has been recommended to prevent droxypregna-4,6-diene-3,20-dione, acetate the estrogen-associated risk of endometrial Synonyms: 17α-Acetoxy-6-chloro-4,6- . Evidence from the Women’s Health -3,20-dione; 6-chloro-Δ6-17- Initiative (WHI) of adverse effects from the use acetoxyprogesterone; 6-chloro-Δ6-[17α] of a continuous combined estrogen–progestogen acetoxyprogesterone has affected prescribing. Patterns of exposure Structural and molecular formulae, and relative are also changing rapidly as the use of hormonal molecular mass therapy declines, the indications are restricted, O CH and the duration of the therapy is reduced (IARC, 3 C 2007). CH3 CH3 O C 1.1 Identification of the agents CH3 H O

1.1.1 Estrogens HH For Estrogens, see the Monograph on O Estrogen-only Menopausal Therapy in this Cl volume.

C23H29ClO4 Relative molecular mass: 404.9

249 IARC MONOGRAPHS – 100A

(b) acetate Structural and molecular formulae, and relative Chem. Abstr. Serv. Reg. No.: 427-51-0 molecular mass H C Chem. Abstr. Name: (1β,2β)-17- 3 OH ′ CH2 (Acetyloxy)-6-chloro-1,2-dihydro-3 H- H C cyclopropa[1,2]pregna-1,4,6-triene-3,20- 2 C CH dione H H IUPAC Systematic Name: 6-Chloro-1β,2β- dihydro-17-hydroxy-3′H-cyclopropa[1,2] HH pregna-1,4,6-triene-3,20-dione acetate Synonyms: Cyproterone 17-O-acetate; cyproterone 17α-acetate; 1,2α-methylene- C H O 6-chloro-17α-acetoxy-4,6-pregnadiene- 22 30 Relative molecular mass: 310.5 3,20-dione; 1,2α-methylene-6-chloro-Δ4,6- pregnadien-17α-ol-3,20-dione acetate; (d) 1,2α-methylene-6-chloro-pregna-4,6- diene-3,20-dione 17α-acetate; methylene- Chem. Abst. Services Reg. No.: 67392-87-4 6-chloro-17-hydroxy-1α,2α-pregna-4,6- Chem. Abstr. Name: (2′S,6R,7R,8R,9S,10R,- diene-3,20-dione acetate 13S,14S,15S,16S)-1,3′,4′,6,7,8,9,10,11,- 12,13,14,15,16,20,21-Hexadecahydro Structural and molecular formulae, and relative -10,13-dimethyl-spiro[17H-dicyclop molecular mass ropa[6,7:15,16]cyclopenta[a]phenan- O CH3 threne-17,2′ (5′H)-furan]-3,5′ (2H)-dione C CH3 CH3 Synonyms: Dihydrospirorenone; 1,2-di- O C hydrospirorenone; drospirenona; spiro[17H-dicyclopropa[6,7:15,16] CH3 H O cyclopenta[a]phenanthrene-17,2′(5′H)- furan]-3,5′(2H)-dione, 1,3′,4′,6,7,8,9,10,11 HH ,12,13,14,15,16,20,21-hexadecahydro-10,13- O dimethyl-, [6R-(6α,7α,8β,9α,10β,13β,14α,1 Cl 5α,16α,17β)]- Structural and molecular formulae, and relative C H ClO 24 29 4 molecular mass Relative molecular mass: 416.9 O (c) H C Chem. Abstr. Serv. Reg. No.: 54024-22-5 3 O Chem. Abstr. Name: (17α)-13-Ethyl-11- methylene-18,19-dinorpregn-4-en-20-yn- CH3 H 17-ol IUPAC Systematic Name: 13-Ethyl-11- HH methylene-18,19-dinor-17α-pregn-4-en- 20-yn-17-ol O Synonyms: 13-Ethyl-11-methylene- 18,19-dinor-17α-4-pregnen-20-yn-17-ol; C24H30O3 17α-ethynyl-18-methyl-11-methylene-Δ4- Relative molecular mass: 366.5 oestren-17β-ol

250 Combined estrogen-progestogen menopausal therapy

(e) (g) Chem. Abstr. Serv. Reg. No.: 152-62-5 Chem. Abstr. Serv. Reg. No.: 60282-87-3 Chem. Abstr. Name: (9β,10α)-Pregna-4,6- Deleted CAS Reg. No.: 110541-55-4 diene-3,20-dione Chem. Abstr. Name: (17α)-13-Ethyl-17- IUPAC Systematic Name: 10α-Pregna-4,6- hydroxy-18,19-dinorpregna-4,15-dien-20- diene-3,20-dione yn-3-one Synonyms: 10α-Isopregnenone; dehydro- IUPAC Systematic Name: 13-Ethyl-17- ; dehydroprogesterone hydroxy-18,19-dinor-17α-pregna-4,15- Structural and molecular formulae, and relative dien-20-yn-3-one molecular mass Structural and molecular formulae, and relative O molecular mass CH3 C H 3C CH3 OH CH2 C CH CH H 3 H H

H H H H O O

C21H28O2 C21H26O2 Relative molecular mass: 312.5 Relative molecular mass: 310.4 (f) Ethynodiol diacetate (h) Chem. Abstr. Serv. Reg. No.: 297-76-7 Chem. Abstr. Name: (3β,17α)-19-Norpregn- Chem. Abstr. Serv. Reg. No.: 797-63-7 4-en-20-yne-3,17-diol, diacetate Deleted CAS Reg. No.: 797-62-6; 4222-79-1; IUPAC Systematic Name: 19-Nor-17α- 121714-72-5 pregn-4-en-20-yne-3β,17β-diol, diacetate Chem. Abstr. Name: (17α)-13-Ethyl-17- Synonyms: Ethinodiol diacetate; ethynodiol hydroxy-18,19-dinorpregn-4-en-20-yn-3- acetate; β-ethynodiol diacetate one IUPAC Systematic Name: 13-Ethyl-17- Structural and molecular formulae, and relative hydroxy-18,19-dinor-17α-pregn-4-en-20- molecular mass yn-3-one O Synonyms: 13-Ethyl-17-ethynyl-

O C CH3 17β-hydroxy-4-gonen-3-one; CH3 C CH 13-ethyl-17α-ethynyl-17- hydroxygon-4-en-3-one; 13-ethyl-17α- H H ethynylgon-4-en-17β-ol-3-one; 13β-ethyl-

H H 17α-ethynyl-17β-hydroxygon-4-en-3-one; O 13-ethyl-17-hydroxy-18,19-dinor-17α- C CH O 3 H pregn-4-en-20-yn-3-one; 17-ethynyl- 18-methyl-19-nortestosterone; 18-methyl-

C24H32O4 norethindrone; l-; dl-norgestrel; Relative molecular mass: 384.5 d-norgestrel

251 IARC MONOGRAPHS – 100A

Structural and molecular formulae, and relative (j) acetate molecular mass Chem. Abstr. Serv. Reg. No.: 71-58-9 H C Chem. Abstr. Name: (6α)-17-(Acetyloxy)-6- 3 OH CH2 methylpregn-4-ene-3,20-dione C CH IUPAC Systematic Name: 17-Hydroxy-6α- H H methylpregn-4-ene-3,20-dione, acetate Synonyms: 17α-Acetoxy-6α- H H methylprogesterone; depomedroxypro- gesterone acetate; depo-progestin; depot- O medroxyprogesterone acetate; DMPA; C H O 17-hydroxy-6α-methylprogesterone, ac- 21 28 2 etate; 17α-hydroxy-6α-methylprogesterone Relative molecular mass: 312.5 acetate; MAP; MPA; medroxypro- (i) gesterone 17-acetate; 6α-methyl-17- acetoxyprogesterone; 6α-methyl-17α- Chem. Abstr. Serv. Reg. No.: 52-76-6 hydroxyprogesterone acetate Deleted CAS Reg. No.: 60416-16-2 Chem. Abstr. Name: (17α)-19-Norpregn-4- Structural and molecular formulae, and relative en-20-yn-17-ol molecular mass

IUPAC Systematic Name: 19-Nor-17α- O CH3 C pregn-4-en-20-yn-17-ol CH CH 4 3 3 Synonyms: 3-Desoxynorlutin; Δ - O C 17α-ethinylestren-17β-ol; Δ4-17α- CH H ethinyloestren-17β-ol; ethynylestrenol; 3 O ethynyloestrenol; 17α-ethynylestrenol; 17α-ethynyloestrenol; 17α-ethynyl-17β- H H 4 hydroxy-Δ -estrene; 17α-ethynyl-17β- O hydroxy-Δ4-oestrene H CH3 Structural and molecular formulae, and relative C H O molecular mass 24 34 4 Relative molecular mass: 386.5 OH CH3 (k) acetate C CH Chem. Abstr. Serv. Reg. No.: 595-33-5 H H Chem. Abstr. Name: 17-(Acetyloxy)- 6-methylpregna-4,6-diene-3,20-dione H H IUPAC Systematic Name: 17-Hydroxy- 6-methylpregna-4,6-diene-3,20-dione, C H O acetate 20 28 Synonyms: DMAP; megestryl acetate; MGA Relative molecular mass: 284.4

252 Combined estrogen-progestogen menopausal therapy

Structural and molecular formulae, and relative (m) acetate molecular mass Chem. Abstr. Serv. Reg. No.: 51-98-9 O CH3 Chem. Abstr. Name: (17α)-17-(Acetyloxy)- C 19-norpregn-4-en-20-yn-3-one CH3 CH3 O C IUPAC Systematic Name: 17-Hydroxy-19- nor-17α-pregn-4-en-20-yn-3-one, acetate CH H 3 O Synonyms: 17α-Ethinyl-19-nortestosterone 17β-acetate; 17α-ethinyl-19-nortestosterone H H acetate; 17α-ethynyl-19-nortestosterone O acetate; norethindrone acetate; norethin- drone 17-acetate; norethisteron acetate; CH 3 norethisterone 17-acetate; 19-norethister- one acetate; norethynyltestosterone acetate; C24H32O4 Relative molecular mass: 384.5 19-norethynyltestosterone acetate; nor- ethysterone acetate (l) Norethisterone Structural and molecular formulae, and relative Chem. Abstr. Serv. Reg. No.: 68-22-4 molecular mass Chem. Abstr. Name: (17α)-17-Hydroxy-19- O norpregn-4-en-20-yn-3-one O C CH3 IUPAC Systematic Name: 17-Hydroxy-19- CH3 nor-17α-pregn-4-en-20-yn-3-one C CH Synonyms: Ethinylnortestosterone; H H 17α-ethinyl-19-nortestosterone; ethynyl- nortestosterone; 17-ethynyl-19-nortestos- H H terone; 17α-ethynyl-19-nortestosterone; O norethindrone; norethisteron; norethyno- drone; 19-nor-17α-ethynyltestosterone; C H O norpregneninolone 22 28 3 Relative molecular mass: 340.5 Structural and molecular formulae, and relative molecular mass (n)

OH Chem. Abstr. Serv. Reg. No.: 3836-23-5 CH3 Chem. Abstr. Name: (17α)-17-(Heptanoyl)- C CH 19-norpregn-4-en-20-yn-3-one H H IUPAC Systematic Name: 17-Hydroxy-19- nor-17α-pregn-4-en-20-yn-3-one, heptano- H H ate O Synonyms: Norethindrone enanthate; norethindrone oenanthate; norethister-

C20H26O2 one enanthate; norethisterone heptanoate; Relative molecular mass: 298.4 17β-hydroxy-19-nor-17α-pregn-4-en-20- yn-3-one heptanoate

253 IARC MONOGRAPHS – 100A

Structural and molecular formulae, and relative Structural and molecular formulae, and relative molecular mass molecular mass O O

H 3C O C (CH2)5 CH3 O C CH3 CH3 CH2 C CH C CH H H H H

H H H H O HO N

C27H38O3 C23H31NO3 Relative molecular mass: 410.6 Relative molecular mass: 369.5 (o) Norethynodrel (q) Norgestrel Chem. Abstr. Serv. Reg. No.: 6533-00-2 Chem. Abstr. Serv. Reg. No.: 68-23-5 Chem. Abstr. Name: (17α)-dl-13-Ethyl-17- Chem. Abstr. Name: (17α)-17-Hydroxy-19- hydroxy-18,19-dinorpregn-4-en-20-yn-3- norpregn-5(10)-en-20-yn-3-one one IUPAC Systematic Name: 17-Hydroxy-19- IUPAC Systematic Name: dl-13-Ethyl-17- nor-17α-pregn-5(10)-en-20-yn-3-one hydroxy-18,19-dinor-17α-pregn-4-en-20- Synonyms: Enidrel; yn-3-one Structural and molecular formulae, and relative Synonyms: (17α)-13-Ethyl-17-hydroxy- molecular mass 18,19-dinorpregn-4-en-20-yn-3-one; OH methylnorethindrone; α-norgestrel; dl- CH3 norgestrel; dl-norgestrel C CH Structural and molecular formulae, and relative H molecular mass H C 3 OH H H CH2 C CH O H H

C20H26O2 Relative molecular mass: 298.4 H H

(p) O

Chem. Abstr. Serv. Reg. No.: 35189-28-7 C H O Chem. Abstr. Name: (17α)-17-(Acetyloxy)- 21 28 2 Relative molecular mass: 312.5 13-ethyl-18,19-dinorpregn-4-en-20-yn-3- one, 3- (r) IUPAC Systematic Name: 13-Ethyl-17- Chem. Abst. Services Reg. No.: 57-83-0 hydroxy-18,19-dinor-17α-pregn-4-en-20- Chem. Abstr. Name: Pregn-4-ene-3,20- yn-3-one oxime acetate () dione Synonyms: 17α-Acetoxy-13-ethyl-17- Synonyms: ; ethynylgon-4-en-3-one oxime; dexnorg- luteal hormone; luteine; luteohormone; Δ4- estrel acetime pregnene-3,20-dione

254 Combined estrogen-progestogen menopausal therapy

Structural and molecular formulae, and relative Therapy in this volume), and ethi- molecular mass nylestradiol. A range of 3–5 different doses are

O CH3 often available for each product, varying from C low doses (0.3–0.5 mg orally) to higher doses CH 3 (2.5–5 mg). The doses of estrogens used are H generally lower than those used in combined CH3 H oral contraceptives, and do not therefore provide contraception. They are available as oral tablets, H H intranasal sprays, skin patches and O , subcutaneous implants, topical applications for vulvovaginal use, and intravaginal rings.

C21H30O2 Generally, if prolonged therapy (for more Relative molecular mass: 314.5 than 2–4 weeks) with an estrogen by any route is envisaged in a woman with an intact , 1.2 Use of the agents a progestogen is given to prevent endometrial proliferation. A range of progestogens are avail- Information for Section 1.2 is taken from able for use in combined therapy. Those most IARC (2007), McEvoy (2007), and Sweetman commonly used are medroxyprogesterone (2008). acetate, norethisterone, levonorgestrel, and micronized progestogens. Several doses of each 1.2.1 Indications progestogen are usually available. While oral forms predominate, progestogens are also avail- Estrogen–progestogen combinations are used able as a vaginal pessary, a systemically absorbed for the treatment of moderate-to-severe vaso- vaginal , a and an intrau- motor symptoms, vulvar and vaginal terine device. The administration of progesto- associated with , and for the preven- gens may follow one of three types of schedule. In tion and treatment of . Women with continuous combined therapy, the same dose of an intact uterus are prescribed a progestogen in both estrogen and progestogen is administered addition to estrogen to reduce the increased risk each day. Cyclic or sequential therapy consists of endometrial carcinoma. of estrogen-alone daily, followed by progestogen with estrogen for 7–20 days and then 5–7 days 1.2.2 Dosages and formulations with no . The duration of each phase can vary. A variety of products are available for use Combined oral products contain both in combined estrogen–progestogen meno- estrogen and progestogen. The various prepara- pausal therapy, either as individual estrogen and tions available differ in their estrogen compo- progestogen components that can be co-admin- nent, their progestogen component, the dose of istered, or as a combined . these components, and the schedule and mode Available products can be defined by their of drug administration. Continuous exposure to estrogen form, dose, and mode of delivery. The both hormones (both estrogen and progestogen most common estrogens available for meno- at fixed daily doses) is common, particularly pausal therapy are conjugated , in the of America, whereas cyclic conjugated plant-based estrogens (A and B; see dosing, in which progestogen is added peri- the Monograph on Estrogen-only Menopausal odically to daily estrogen, is prevalent in other

255 IARC MONOGRAPHS – 100A countries. Other scheduling strategies are also developing countries. At the peak of use in 1999, used occasionally. approximately 20 million women in developed When (A or B) are countries used combined hormonal therapy. used with medroxyprogesterone acetate for Use has fallen by more than 50% since 2002, the management of moderate-to-severe vaso- particularly for continuous combined hormonal motor symptoms associated with menopause therapy. Use in some developing countries has or for the management of vulvar and vaginal also declined modestly, although the data are atrophy, conjugated estrogens are administered more limited. Among peri- and postmenopausal in a continuous daily dosage regimen while women in developed countries, current users of medroxyprogesterone acetate is administered in combined hormonal therapy tend to be younger a continuous daily dosage regimen or cyclically. and more highly educated, to have a lower body When both drugs are administered in a contin- mass, and to use health care more regularly than uous daily dosage regimen, conjugated estrogens non-users (IARC, 2007). are administered in a daily dosage of 0.3 mg in conjunction with oral medroxyprogesterone acetate in a daily dosage of 1.5 mg. Alternatively, 2. Cancer in Humans conjugated estrogens are administered in a daily dosage of 0.45 mg in conjunction with medroxy- 2.1 Cancer of the progesterone acetate in a daily dosage of 1.5 mg, or conjugated estrogens are administered in a At the time of the first IARC Monograph daily dosage of 0.625 mg in conjunction with of hormones and (IARC, 1999), medroxyprogesterone acetate in a daily dosage almost all of the epidemiological evidence came of 2.5 or 5 mg. When conjugated estrogens from studies that evaluated estrogen prescribed are administered in a continuous daily dosage without a progestogen. Data on breast cancer regimen and medroxyprogesterone acetate is risk related to estrogen plus a progestogen were administered cyclically, conjugated estrogens are deemed insufficient to reach any firm conclusions administered in a daily dosage of 0.625 mg, while about the carcinogenicity of combined hormone oral medroxyprogesterone acetate is adminis- therapy. tered in a daily dosage of 5 mg on Days 15–28 of The next IARC Monograph on this topic the cycle. (IARC, 2007) reviewed two randomized trials, ten cohort studies, and seven case–control studies on 1.2.3 Trends in use combined postmenopausal and breast cancer published up to and including 2004. When the addition of a progestogen to estro- These studies consistently reported an increased gens was introduced after 1975 as a strategy to risk for breast cancer in users of estrogen plus reduce the risk of , the use progestogen therapy compared with non-users. of the combination for menopausal therapy The risk increased with increasing duration of use, increased steadily in the 1980s, particularly was largely confined to current or recent users, in developed countries. Combined estrogen– and decreased soon after hormone treatment was progestogen menopausal therapy is now admin- stopped. Although the previous IARC evalua- istered to women who have not undergone a tion concluded that there is sufficient evidence . in humans for the carcinogenicity of combined Combined hormonal therapy is much more estrogen–progestogen menopausal therapy in commonly used in developed countries than in the breast, it was not possible to evaluate whether

256 Combined estrogen-progestogen menopausal therapy breast cancer risk varied according to the type analysis of ten studies comparing continuous of progestogen or its dose, or according to the combined progestogens with sequential progesto- number of days each month that the progestogen gens, half of the studies suggested a higher risk was taken. with continuous combined therapy, but most The present review of studies published differences were small and had overlapping 95% through August 2008 includes four new system- confidence intervals (see Table 2.1 available at atic reviews, additional analyses from two http://monographs.iarc.fr/ENG/Monographs/ clinical trials, five cohort studies, and four case– vol100A/100A-13-Table2.1.pdf). control studies, as well as many studies of time Another meta-analysis of papers published trends, and two trials of hormone therapy in between 1989–2004 included 21 case–control breast cancer survivors. Studies were included if studies, 15 cohort studies, and six controlled the authors provided risk estimates (odds ratios clinical trials. Trials and cohort studies had [OR], hazard ratios [HR] or relative risks [RR]) more cancer cases than case–control studies, but and 95% confidence intervals (CI) comparing provided separate stratified analyses of estrogen breast cancer risk in women who used combined plus progestogen for only three ever-never use (estrogen plus progestogen) hormone therapy comparisons, and four duration-of-use anal- with non-hormone users for at least 1 year, or if yses. Most of the relevant results are from the the authors specified that at least 80% of estrogen- subset of ten case–control studies that reported taking women were likely to be using combined combined hormone treatment separately. The therapy. Evidence from many studies of varying summary statistics showed an increased cancer breast cancer during recent years of risk for estrogen plus progestogen, especially for increasing and then decreasing prescription data after 1992, when estrogen plus progestogen of combined hormone therapy is also reviewed. in a single tablet became more widely available (Greiser et al., 2005). 2.1.1 Systematic reviews Shah et al., (2005) published a meta-analysis of eight studies of current hormone therapy and Four systematic reviews published after 2004 breast cancer excluding women with a history comprise studies of combined postmenopausal of oral contraceptive use. The summary analysis hormone use and breast cancer risk (Campagnoli showed that estrogen plus progestogen use for et al., 2005; Collins et al., 2005; Greiser et al., 2005; less than 5 years significantly increased breast Shah et al., 2005). All support the conclusion that cancer risk (OR, 1.35; 95%CI: 1.16–1.57), and use the use of estrogen plus progestogen increases for more than 5 years showed a somewhat greater the risk of breast cancer in women, although risk (OR, 1.63; 95%CI: 1.22–2.18). each review included a somewhat different set of A by Collins et al. (2005) studies, offered some distinct conclusions based included published data on estrogen plus on the different studies chosen, had different progestogen and breast cancer from four rand- definitions of hormone exposure, and different omized trials, two of which – WHI and subset analyses related to different hormone and Estrogen/Progestin Replacement Study regimens, duration, and recency of use. (HERS) – were included in the previous IARC Campagnoli et al. (2005) reviewed several Monograph, and two other small earlier trials publications, mainly those reviewed in the that together added only five cancer cases (all previous IARC Monograph, and confirmed a in the group), which did not change significantly increased risk of breast cancer risk the overall risk estimates. This review includes with combined hormone therapy. In a subset

257 IARC MONOGRAPHS – 100A very useful pre-planned subset analyses with the make it less likely that lack of power concealed following results: clinically meaningful differences.] • the for breast cancer was lower in the intent-to-treat analysis (OR, 2.1.2 Studies of changing breast cancer 1.24; 95%CI: 1.01–1.54) than in the adher- incidence in the context of changing ent women analysis (OR, 1.49; 95%CI: patterns of menopausal hormone 1.13–1.96), compatible with the high therapy use drop-out and crossover rate in the WHI- Estrogen-Progestogen Trial (EPT); The widely publicized increased risk of • the dominance of the Million Women breast cancer in the WHI-EPT trial published Study (Beral et al., 2003) (1934 estrogen in 2002 (Rossouw et al., 2002) was followed by user cases) did not explain the excess rapid and substantial drops in the incidence breast cancer risk associated with com- of invasive breast cancer in the USA (Clarke bined hormone therapy in published et al., 2006; Glass et al., 2007; Jemal et al., 2007; studies, in that the published summary Kerlikowske et al., 2007; Ravdin et al., 2007; risk in the seven epidemiological stud- Robbins & Clarke, 2007), (Katalinic & ies that included the Million Women Rawal, 2008), (Allemand et al., 2008), and Study (OR, 1.70; 95%CI: 1.36–2.13) was Switzerland (Bouchardy et al., 2006). The most very similar to the summary risk for the striking changes were observed in countries with other six studies that did not include the the highest rates of postmenopausal estrogen use Million Women Study (OR, 1.67; 95%CI: before the WHI-EPT trial results were known. 1.29–2.17); Declines were not reported in Norway or Sweden • the analysis confirms the WHI-EPT data (Zahl & Maehlen, 2007) or in African-Americans showing a significantly increased risk of in the USA (Hausauer et al., 2007). Most of the breast cancer begins within 5 years of ini- studies were based on representative regional or tiating combined therapy and increases national cancer registries, with validated diag- with increasing years of use; the analy- noses, but had only self-reported data on indi- sis highlights an important result from vidual hormone treatment or mammography. the Million Women Study, showing A recent review of 21 papers published from similar relative risks for equine estrogen 1987–2007 on the incidence of breast cancer in the (OR, 1.29; 95%CI: 1.16–1.43) and estradiol USA in the 1980s (Krieger, 2008) documented the (OR, 1.24; 95%CI: 1.12–1.37); and, rise in breast cancer in women aged 50 years and • an analysis (based on pooled data from older (9.9% per year in the 1980s) and the decrease the Million Women Study and the (by 50%) since 2002 in the context of changing Danish Nurses Cohort) showed essen- frequency of hormone therapy. In a French study tially identical risk for C21 progestogens (Allemand et al., 2008), the age-adjusted breast (medroxyprogesterone acetate) and C19 cancer incidence increased 2.1% per year during progestogens (norethisterone, levonorg- 2000–04; decreased 4.3% during 2004–05, and estrel) (OR, 2.14 for each). 5.3% during 2005–06, a decrease observed only [The Working Group noted that more than in women aged 50 years and older. This decrease 400 user cases for the estradiol versus equine was not likely to be explained by decreased estrogen comparison and the more than 1900 mammography after cessation of hormones user cases for the progestogen comparisons because the use of mammography increased by 335% during 2000–06.

258 Combined estrogen-progestogen menopausal therapy

The only study of time trends in breast cancer in postmenopausal women using estrogen incidence that provides data on cancer incidence plus progestogen (see Table 2.2 available at by cancer stage and estrogen positive http://monographs.iarc.fr/ENG/Monographs/ (ER+) status, mammography use, and hormone vol100A/100A-13-Table2.2.pdf). prescription rates in the same cohort comes In a cohort study using data from a previ- from a large pre-paid health care plan in the ously published Danish Nurses Study, Stahlberg USA, which included screening mammography, et al. (2004) reported that current estrogen plus pharmacy dispensing of hormone therapy, and cyclical progesterone-like progestogen, cyclical a tumour registry (Glass et al., 2007). The age- -like progestogen, or continuous adjusted incidence rates of breast cancer increased testosterone-like progestogen were each signifi- by 25% from the early 1980s to 1992–93, then cantly associated with an increased risk of ductal increased an additional 15% through 2000–01, carcinoma (OR, 3.10; 95%CI: 1.69–5.67; OR, 2.15; and dropped by 18% in 2003–04. Increases were 95%CI: 1.31–3.54; and OR, 4.10; 95%CI: 2.29– mainly for ER+ breast in women aged 45 7.30, respectively). years and older. During 1993–2006, 75–79% of Fournier et al. (2005) reported the risk women older than 45 years of age were screened of breast cancer in 54548 postmenopausal within the previous 2 years. Postmenopausal French women who were followed an average hormone prescriptions, primarily estrogen plus of 2.8 years postmenopause, during which time progestogen, increased during 1988–2002, and 948 new invasive breast cancers were diag- dropped by approximately 75% after 2002, coin- nosed. There was a similar increased risk of cident with the publication of the WHI-EPT breast cancer with transdermal/percutaneous (estrogen plus progestogen) results (RR, 1.4; 95%CI: 1.2–1.7) and oral estrogens that showed that combined hormone therapy (RR, 1.5; 95%CI: 1.1–1.9) when combined with a increased the risk of heart disease and breast synthetic progestogen, compared to the risk in cancer (Rossouw et al., 2002). non-hormone users. There was no increased risk Overall, the studies of time trends in ER+ when estrogen was combined with micronized breast cancers in women aged 45 or older are progesterone (RR, 0.9; 95%CI: 0.7–1.2) (P test for compatible with a substantial increase in breast heterogeneity < 0.001). cancer risk associated with increasing meno- Lee et al. (2006) reported a prospective study pausal hormone use, with a remarkable decrease of combined hormone therapy and breast cancer concurrent with the release of the WHI-EPT in 55371 African-American, Native Hawaiian, results. Although the increase in breast cancer Japanese-American, Latina, and Caucasian post- could be partly explained by the increasing use menopausal women from the US Multiethnic of mammography in women concordant with Study. The authors provide no information increasing hormone therapy, the decrease in about the type of estrogen or progestogen used, breast cancer concurrent with reduced hormone but combined equine estrogen and medroxypro- was not explained by a decrease in mammography. gesterone acetate accounted for more than 80% of combined hormone prescriptions in the USA 2.1.3 Cohort studies during the mid-and late 1990s. Current use of estrogen plus progestogen was associated with All five cohort studies Stahlberg( a significantly increased risk of breast cancer et al., 2004; Fournier et al., 2005; Lee et al., within the first 5 years of use (adjusted RR, 2006; Rosenberg et al., 2006a; Corrao et al., 1.43; 95%CI: 1.06–1.93), and this risk increased 2008) reported an increased risk of breast cancer with duration of use. The increased risk was

259 IARC MONOGRAPHS – 100A associated with both ductal and lobular cancer, in the trial. Hazard ratios for estrogen plus more advanced cancer (based on standard progestogen users compared to non-users were pathological criteria), and ER+/Progesterone close to 2 (95%CI: 1.86–2.20) in the trial and the Receptor positive (PR+), ER+/PR-, and ER-/PR- groups, except for the clin- tumours. It was statistically significant in all ical trial participants without prior hormone use ethnic groups, and persisted after adjusting for who had a much smaller risk ratio (1.13). Women the lower frequency of mammograms in women who initiated hormone therapy within 5 years of not taking hormones. the menopause had a significantly higher risk of Rosenberg et al. (2006a) reported an 8-year breast cancer than those who initiated hormone follow-up study of 32559 African-American therapy later; and the risk increased with duration women. Compared to non-hormone users, the of use (HR, 1.85; 95%CI: 1.03–3.34) for 2–5 years incidence rate ratio associated with 10 or more of use and (HR, 2.75; 95%CI: 1.73–4.39) for more years of estrogen plus progestogen was 1.45 than 5 years of use. (95%CI: 0.94–2.23). Shorter durations of use were A new analysis of the Million Women Study not associated with estrogen plus progestogen by Reeves et al. (2006) assessed the relative risk therapy in these African-American women. of current hormone therapy for different histo- Corrao et al. (2008) followed 73505 Italian logical types of breast cancer. In analyses of women in Lombardia who had received at least current combined hormone use versus never one prescription for postmenopausal hormone use, the risk was significantly increased for all therapy in 1998–2000, and were followed until types of cancers for which there were more than 2005. More than 88% began treatment using 50 cases, including lobular cancer (n = 503, RR, transdermal estradiol; and combined hormone 2.80; 95%CI: 2.46–3.18), tubular cancer (n = 186, use was assumed because few Italian women have RR, 3.51; 95%CI: 2.80–4.41), and ductal cancer had a hysterectomy. Breast cancer risk increased (n = 2241, RR, 2.0; 95%CI: 1.89–2.12). The risk for with duration of therapy and was greater with these three most common types of breast cancer oral than with transdermal estradiol. The odds increased significantly with increasing duration ratio for at least 25 months of transdermal estra- of therapy. diol was 1.27 (95%CI: 1.07–1.51), compared to 2.14 (95%CI: 1.43–3.21) for oral estradiol (P for 2.1.4 Case–control studies heterogeneity < 0.01). This difference is consistent with and somewhat larger than that reported in Four case–control studies also showed the initial report from the Million Women Study an increased risk of breast cancer in (Beral et al., 2003), with a similar mean follow-up women using estrogen plus progestogen (Li of 2.6 years. et al., 2006; Rosenberg et al., 2006b, c; Wu A new analysis from Prentice et al. (2008) et al., 2007; see Table 2.3 available at http:// combined data from the WHI-EPT trial monographs.iarc.fr/ENG/Monographs/ (n = 16608) and the subset of women from the vol100A/100A-13-Table2.3.pdf). WHI observational study who were either not In a Swedish study (Rosenberg et al., 2006c), taking hormones at baseline (n = 32084) or women who had used a medium- estrogen were users at enrollment of the same hormone (i.e. not ) plus a progestogen for 5 or more regimen used in the WHI-EPT trial (n = 25328). years had a significantly increased risk of lobular Women included from the observational study cancer (OR, 5.6; 95%CI: 3.2–9.7), tubular cancer were also required to have had a mammogram (OR, 6.5, 95%CI: 2.8–14.9), and ductal cancer in the past two years, to parallel the protocol (OR, 2.3; 95%CI: 1.6–3.3). In another report

260 Combined estrogen-progestogen menopausal therapy from the same study (Rosenberg et al., 2006b), cancer if they were former oral contraceptives estrogen (mainly estradiol) plus a progestogen users (RR, 2.45; 95%CI: 1.92–3.12) than if they (mainly levonorgestrel or norethisterone) for at had never used oral contraceptives (RR, 1.67; least 5 years significantly increased the risk of 95%CI: 1.32–2.12) (P = 0.002). The odds ratios ER+/PR+ cancers (OR, 3.0; 95%CI: 2.1–4.1) but in current hormone users who were past oral not ER-/PR- tumours (OR, 1.3; 95%CI: 0.7–2.5). contraceptives users were very similar in women A study of Asians (Chinese, Japanese or currently using estrogen (primarily estradiol) Filipino) living in the USA (Wu et al., 2007) found alone (OR, 2.63; 95%CI: 1.65–4.20) or estrogen a 26% increased risk of breast cancer among plus a progestogen (primarily a testosterone current users of estrogen plus progestogen for derivative) (OR, 2.55; 95%CI: 1.94–3.35). each 5 years of use (OR, 1.26; 95%CI: 1.04–1.52). Li et al. (2006) reported breast cancer 2.1.6 New WHI-EPT clinical trial analyses histology in a multicentre case–control study in the USA. Compared to never use, current use of Anderson et al. (2006) compared breast cancer estrogen plus progestogen was associated with an rates in women in the WHI-EPT trial according increased risk of all types of breast cancer; the to their history of menopausal hormone use excess risk was statistically significant for ductal- before beginning the trial. Despite controlling lobular (OR, 2.9; 95%CI: 1.7–4.9) and tubular for an extensive list of potential confounders, (OR, 3.2; 95%CI: 1.3–7.5) cancers. there was a significantly greater risk of invasive breast cancer in women who reported pre-WHI 2.1.5 Postmenopausal breast cancer risk with postmenopausal hormone use (HR, 1.96; 95%CI: prior oral contraceptive use, hormone 1.17–3.27) compared to the women without such a history (HR, 1.02; 95%CI: 0.77–1.36). These therapy use, or both women also had a significantly higher risk of a Dumeaux et al. (2005) reported breast larger tumour, higher number of positive nodes, cancer risk in current hormone users according and more regional/metastatic disease, but the to prior use of hormones in a French cohort of number of women in these subsets was small 68670 postmenopausal women. The most widely and the 95% confidence intervals were large. The used postmenopausal hormone regimen was authors note that they did not adjust for multiple transdermal estrogen in combination with either comparisons and that there were significant micronized progesterone or a progesterone differences between the women who did or did derivative. In women currently using hormone not use hormones before entering the WHI-EPT therapy, a history of prior oral contraceptive trial, not all of which could be characterized well use did not further increase the risk (OR, 0.91; enough to exclude residual . 95%CI: 0.81–1.03), but women with a history Heiss et al. (2008) examined the effect of stop- of postmenopausal hormone use did have an ping combined hormone treatment 2–3 years increased risk (OR, 1.41; 95%CI: 1.26–1.59). after the WHI-EPT trial was stopped. More inva- Results were similar for women who had used sive breast cancers occurred in women assigned both oral contraceptives and postmenopausal to continuous estrogen and progestogen than hormones (OR, 1.43; 95%CI: 1.25–1.64). women assigned to placebo (HR, 1.27; 95%CI: In contrast, in a Norwegian cohort study of 0.91–1.78), with a risk similar to that observed 30118 postmenopausal women, reported by Lund for combined therapy during the trial (HR, 1.26; et al. (2007), current users of hormone therapy 95%CI: 1.02–1.55). These data are the strongest sustained a significantly greater risk of breast evidence for a continued excess risk 2–3 years

261 IARC MONOGRAPHS – 100A after abrupt cessation of hormone therapy. [The to Stage II), were free of recurrence an average Working Group noted that one limitation of of 3 years after cancer treatment, had no other this study is that women who abruptly stopped serious disease, and had severe menopause hormone therapy at the end of the trial may have symptoms. Women were randomly assigned to restarted within a year when menopause symp- hormones or to the best available symptom treat- toms recurred.] ment without hormones. Hormone treatment in In conclusion, many new studies confirm women without a hysterectomy included either the earlier studies reporting that estrogen plus cyclic or continuous estrogen and progestogen. progestogen increases the risk of breast cancer in At the time the trial was stopped early for postmenopausal women. The more recent studies harm, after an average of 2 years with at least provide additional evidence that the increased one follow-up visit, the for women risk begins within 5 years of initiating combined assigned to any estrogen regimen with or without therapy and increases with increasing duration a progestogen was 3.5 (95%CI: 1.5–8.1). The risk of therapy. The data are insufficient to determine was not changed after adjusting for prior use of whether the risk differs by estrogen type, dose, or hormone replacement therapy, , and or the progestogen type ER positivity. (progesterone or progesterone-derived versus In a follow-up study, Holmberg et al. (2008) testosterone derived) or regimen (continuous showed persistence of harm after an extended versus cyclic) progestogen. The data provide no 4-year follow-up of 442 of the 447 women in consistent evidence that any histological type of the HABITS trial; 39 of the 221 women in the cancer is more often associated with combined hormone-treated group and 17 of the 221 women hormone therapy in postmenopausal women. in the untreated group had experienced recurrent breast cancer (HR, 2.4; 95%CI: 1.3–4.2), i.e. 19 2.1.7 Postmenopausal hormone therapy additional cases since the trial had been stopped. after breast cancer Compared to the 100 women taking estrogen plus continuous progestogen, the hazard ratio Col et al. (2005) published a meta-analysis for recurrence was increased in women taking of two uncontrolled and unblinded randomized different types of hormone therapy (HR, 1.4 for trials and eight observational studies of breast the 150 women taking estrogen plus a continuous cancer survivors, of whom 1316 reported unspeci- progestogen; 1.4 for the 100 taking sequential fied hormone therapy and 2839 did not. The trials progestogen, and 1.4 for the 53 women taking showed that hormone therapy increased breast progestogen, primarily ), cancer recurrence (RR, 3.41; 95%CI: 1.59–7.33), but numbers in each group were small and confi- while the observational studies suggested it dence intervals were wide. reduced risk (RR, 0.64; 95%CI: 0.50–0.82). The Concurrent with the HABITS trial, the apparently protective associations in observa- Stockholm randomized clinical trial also tional studies were attributed by the authors enrolled Swedish women with a history of to probable selection for hormone treatment of breast cancer (von Schoultz & Rutqvist, 2005). younger, more often node-negative women. The authors reported that the 77% of women Holmberg & Anderson (2004) reported who had had a hysterectomy were treated with preliminary results from HABITS, a randomized , and 23% with estradiol plus clinical trial of hormone therapy for menopause cyclic or spaced low-dose medroxyprogesterone symptoms in Scandinavian patients aged 40–70 acetate. After a median follow up of 4.1 years, years old who had a history of breast cancer (up there were 11 breast cancer recurrences in

262 Combined estrogen-progestogen menopausal therapy the hormone-treatment group and 13 in the are from the USA, where more than 80% of control group (HR, 0.82; 95%CI: 0.35–1.9). The combined estrogen regimens were conjugated authors speculated that the absence of harm in equine estrogen plus medroxyprogesterone the Stockholm trial could reflect the low dose acetate. The newer studies provide contradic- of medroxyprogesterone instead of the nore- tory evidence about the minimum progestogen thisterone acetate progestogen used in HABITS, required to reduce the estrogen-induced risk of but they also report that the women in this trial endometrial cancer. had less node-positive cancer, and were more In a population-based case–control study likely to have had prior treatment with tamoxifen in Pennsylvania, Strom et al. (2006) compared than women in the HABITS trial. 511 endometrial cancer cases detected by active In summary, there is consistent evidence that surveillance of regional hospitals with 1412 combined estrogen–progestogen menopausal controls of similar age identified mainly from therapy increases the risk of breast cancer. There random-digit dialling. The history of hormone is evidence for an increasing risk with increasing use was determined mainly by telephone, duration of use among current users. However, using a structured questionnaire and memory determining whether all current formulations aids mailed in advance. Use of combined and treatment regimens are equally carcinogenic hormones of any duration was not associated is not possible on the available data. with an increased risk of endometrial cancer (OR, 0.8; 95%CI: 0.6–1.1) (see Table 2.4 available 2.2 Cancer of the at http://monographs.iarc.fr/ENG/Monographs/ vol100A/100A-13-Table2.4.pdf). Numbers were The previous IARC Monograph (IARC, too small for useful comparisons of continuous 2007) concluded that there is sufficient evidence and sequential progestogen therapy. that estrogen plus progestogen for at least 14 Using data from three population-based days can prevent estrogen-induced endometrial case–control studies in different counties of cancer. The most compelling evidence was from Washington state, Weiss et al. (2006) examined the Million Women Study (Beral et al., 2005), the aggressiveness of endometrial cancer (based and from the WHI-EPT trial (Anderson et al., on pathology review) in 1304 cases from the state 2003), where endometrial cancer rates were low cancer registry, and 1779 controls of similar age and were not increased by 5 years of continuous who were recruited by random-digit dialling. combined estrogen plus progestogen in a single Combined hormone therapy was not signifi- tablet. The Million Women Study (1320 endo- cantly associated with aggressive endometrial metrial cancer cases) showed that the protective cancer (OR, 1.6; 95%CI: 0.4–7.2). The risk for the effect of a progestogen added to daily estrogen least aggressive endometrial cancer was increased increased with increasing number of days/month with combined therapy when the progestogen that progestogen was used, while the WHI-EPT was used for less than 10 days a month for at results (58 endometrial cancer cases) related least 4 years (OR, 6.2; 95%CI: 3.2–12.0) and for only to daily (continuous) combined hormone 10–24 days/month for at least 4 years (OR, 2.9; therapy and only to one progestogen, medroxy- 95%CI: 1.6–5.0). progesterone acetate. In a cohort study of 30379 women from the Since the last IARC Monograph, two new US National Cancer Institute Breast Cancer case–control studies and two cohort studies have Detection Demonstration Project, started been published on the association of estrogen in 1979 with a 13-year follow-up, combined plus progestogen with endometrial cancer. All hormone use was first queried in 1987–89, and

263 IARC MONOGRAPHS – 100A periodically thereafter by mail and telephone of endometrial cancer is inversely associated with (Lacey et al., 2005). Cancer was identified from the number of days per month that progestogens self-reports, medical records, cancer registries, are allied to the regimen. It is not known whether and certificates. Endometrial cancer risk continuous use (or daily use) reduces the risk of was increased with exclusive use of estrogen plus endometrial cancer compared to baseline. progestogen for less than 15 days/month (RR, 3.0; 95%CI: 2.0–4.6, 32 cases) and for more than 15 2.3 Cancer of the colorectum days/month (RR, 2.3; 95%CI: 1.3–4.0, 15 cases) (see Table 2.5 available at http://monographs. At the time of the previous IARC Monograph iarc.fr/ENG/Monographs/vol100A/100A-13- (IARC, 2007), data from two randomized trials, Table2.5.pdf). Risk increased with increasing four cohort studies, and three case–control duration of use for both regimens. studies showed no elevated risks of colorectal A cohort study of 73211 women from the cancer in women taking combined postmeno- National Institutes of Health-AARP Diet and pausal hormones. In fact, all but one study Health Study included 51312 women who either showed a relative risk estimate less than one never used estrogen or only used estrogen plus (with a statistically significant reduction in two progestogen for at least 10 days/cycle (Lacey et al., of these) suggesting protection. The reduced risk 2007). Hormone use was self-reported by mail, was mainly in recent users and unrelated to dura- and cases were ascertained from state cancer tion of use. Since the last review, there have been registries and death indices. Compared to non- two new case–control studies and one cohort hormone users, neither estrogen plus progestogen study that included analyses largely restricted for 10–14 days/cycle (RR, 0.74; 95%CI: 0.39–1.40, to (or almost entirely restricted to) estrogen plus based on 11 cases) or for at least 20 days/cycle progestogen regimens. (RR, 0.80; 95%CI: 0.55–1.15, based on 35 cases) A US population-based case–control study was associated with an increased risk of endome- conducted in Washington state of 578 cancer trial cancer. Similar results were seen in analyses cases and 590 controls, in which a history of restricted to women who had used combined hormone use and covariates was based on a therapy with either regimen for at least 3 years. 60-minute telephone interview, showed a 40% In conclusion, three of the four new US studies reduced risk of (OR, 0.60; do not show consistent prevention of estrogen- 95%CI: 0.05–0.09) in women who used combined associated endometrial cancer risk in women hormone therapy exclusively for at least 5 years taking estrogen plus sequential progestogen compared to non-users (Newcomb et al., 2007) (mainly medroxyprogesterone acetate) for at (see Table 2.6 available at http://monographs. least 15 days a month, and contradict findings iarc.fr/ ENG/Monographs/vol100A/100A-13- reported in many earlier studies (Anderson et al., Table2.6.pdf ). The reduced risk was restricted 2003; Beral et al., 2005). [The Working Group to current users. There was no association with noted that the reason for these differences is not cancer stage at diagnosis. clear, but poor hormone regimen recall or adher- A German case–control study of 354 cases and ence cannot be excluded.] 1422 age-matched controls (Dinger et al., 2007) There is consistent evidence that the risk of showed no significant associations with colo- endometrial cancer is increased in women taking rectal cancer by progestogen type (medroxypro- unopposed estrogen, and the increased risk gesterone use was rare), duration of progestogen remains evident when the opposing progestogen use, or sequential versus continuous progestogen is taken for less than 15 days per month. The risk

264 Combined estrogen-progestogen menopausal therapy use. The number of cancer cases in each subset cancer (RR, 1.58; 95%CI: 0.77–3.24) (Anderson analysis was small. et al., 2003). A cohort study from Lund, Sweden (Nazeri Since the previous IARC Monograph, there et al., 2006) included 2452 women who reported have been two new case–control studies, three estrogen plus progestin therapy, primarily a new cohort studies, and one new meta-analysis fixed-dose preparation containing 2 mg of on estrogen plus progestogen therapy and risk of estradiol and 2 mg of norethisterone, and 3600 . women not taking postmenopausal hormones. Moorman et al. (2005) reported a North There was a reduced risk of colorectal cancer in Carolina (USA) population-based case–control women taking combined hormones (OR, 0.18; study of ovarian cancer with 364 cases and 370 95%CI: 0.04–0.84) based on only 16 cases in the controls in postmenopausal women (see Table 2.8 no-hormone group and two cases in the women available at http://monographs.iarc.fr/ ENG/ reporting mostly combined hormone therapy Monographs/vol100A/100A-13-Table2.8.pdf). (see Table 2.7 available at http://monographs. Exposure to hormone therapy and covariates iarc.fr/ ENG/Monographs/vol100A/100A-13- was obtained by a 90-minute interview. Sources Table2.7.pdf ). Information about the type of of cancer cases included a rapid case ascertain- estrogen or progestogen was not provided. ment system and a state-wide cancer registry [The Working Group noted that one of two (only 70 cases and 87 controls used combined case–control studies and a single cohort study estrogen and progestogen exclusively). The only found that combined hormone therapy reduced significant association was in women who had the risk of colorectal cancer, which is compatible used combined hormones, but not exclusively, with suggestive evidence in the previous IARC for > 119 months (22 cases, 14 controls); these Monograph. One of these three studies found women had an of 2.4 (95%CI: 1.1–5.3). that reduced risk was seen only with current Types of cancer did not differ by hormone use hormone use of combined hormone therapy. history. These results are suggestive but insufficient to Rossing et al. (2007) reported on a US case– conclude that estrogen plus progestogen therapy control study of 1054 women (440 cases, 614 protects against colorectal cancer, or to conclude controls) in western Washington state. Hormone that this putative effect varies by type of estrogen exposure was determined by interview, aided by or progestogen used.] photographs to identify pills. Cancer data were from the regional cancer registry. No increased 2.4 Cancer of the risk was reported among current users who used only estrogen plus progestogen, regardless of In the previous IARC Monograph (IARC, duration of use (OR, 1.1; 95%CI: 0.8–1.5). 2007), results from two randomized trials, four The Million Women Study of postmeno- cohort studies, and three case–control studies pausal women is the most compelling new cohort were inadequate to establish an association study based on its size, prospective design, large between ovarian cancer and combined estrogen– number of ovarian cancer cases, data on histo- progestogen therapy. The largest clinical trial logical subtypes, high proportion of women (WHI), in which 16608 women were assigned taking hormone therapy (30%), and data for the to conjugated equine estrogen and continuous most important covariates including hysterec- medroxyprogesterone acetate or placebo, did tomy (Beral et al., 2007). In analyses limited to not show a significantly increased risk of ovarian estrogen plus progestogen use (different strati- fied analyses included 69263 cancer cases), the

265 IARC MONOGRAPHS – 100A risk of ovarian cancer was significantly increased A meta-analysis of 42 studies of hormone only in women who had taken combined therapy and ovarian cancer published from hormones for at least 5 years (RR, 1.53; 95%CI: 1966–2006 included 30 case–control studies, 1.27–1.84), and was similar for continuous or seven cohort studies, one randomized clinical combined regimens. The risk was significantly trial and four cancer registry studies, with 12238 increased for estrogen plus norethisterone, non- ovarian cancer cases. The summary risk for ever significantly increased for norgestrel, and not versus never estrogen plus progestin use, based increased for medroxyprogesterone acetate. The on 31 data sets but no explicitly stated number of overall increased risk of hormone therapy was cancer cases, was 1.11 (95%CI: 1.02–1.21). There primarily for epithelial serous tumours; histolog- was no evidence for a significantly increased risk ical data were not shown stratified by combined per year of use (22 data sets). Risks were slightly hormone therapy (see Table 2.9 available at greater for European women (OR, 1.06; 95%CI: http://monographs.iarc.fr/ ENG/Monographs/ 1.03–1.09, based on 14 data sets) than for North vol100A/100A-13-Table2.9.pdf). American women (OR, 1.00; 95%CI: 0.96–1.04, A second cohort study from the NIH-AARP based on seven data sets). Within the estrogen Diet and Health Study (Lacey et al., 2006), plus progestogen data sets, there were no differ- which included 97638 postmenopausal women ences by histological subtypes of ovarian cancer. (of whom 73483 did not undergo a hysterec- Funnel plots suggest publication bias (not tomy, and 51698 had used no hormone therapy publishing small studies showing increased or only estrogen plus progestogen), reported a cancer risk) (Greiser et al., 2007). significantly increased risk associated with 5 The overall risk estimate in a recent meta- or more years of estrogen plus progestogen use analysis was 1.1 for ever versus never use. The two either sequentially (RR, 3.09; 95%CI: 1.68–5.68) new case–control studies and three new cohort or continuously (RR, 1.82; 95%CI: 1.03–3.23). studies of ovarian cancer and combined hormone The US Nurses Health Study followed 7394 therapy suggest only very small increased ovarian postmenopausal women who had at some point cancer risk after use for 5 or more years; however, reported current use of combined estrogen plus the evidence is not consistent across studies. progestogen and 20853 never users (Danforth et al., 2007) (the cohort was followed from 2.5 Cancer of the skin 1976–2002, but few women were using estrogen plus progestin until the 1980s, and the duration Since the previous IARC Monograph (IARC, of relevant follow-up and the number of cases for 2007), an updated analysis of data from a hospital- different analyses are not clear). No significant based case–control study in San Francisco, USA association between estrogen plus progestogen (Lea et al., 2007), and a report on a hospital-based use and epithelial ovarian cancer was observed case-control study in (Naldi et al., 2005) (RR, 1.04; 95%CI: 0.82–1.32, based on 82 cases), or have been published. These were based on 318 by duration or recency of use, but there were only cases and 395 frequency-matched controls, and ten ovarian cancer cases among women using 316 cases and 308 controls, respectively. Neither estrogen plus progestogen exclusively. There was studies showed any association between meno- also no association with serous (RR, 1.12; 95%CI: pausal therapy and risk of cutaneous melanoma. 0.84–1.51, based on 49 cases) or endometrioid tumours (RR, 1.04; 95%CI: 0.53–2.03, based on 15 cases).

266 Combined estrogen-progestogen menopausal therapy

2.6 Cancer of the 2.8 Cancers of the central nervous system Since the previous IARC Monograph (IARC, 2007), results from a population-based case– In the cohort study based on the Canadian control study in New Caledonia, France, an area National Breast Screening study (Silvera et al., with an unusually high incidence of thyroid 2006), 59 incident glioma cases occurred during cancer, have been published (Truong et al., 2005). an average 16.4 years of follow-up in postmen- Answers to in-person interviews of 293 cases opausal women. Based on answers to a self- and 354 controls selected from electoral rolls administered questionnaire at recruitment into were compared. The odds ratio in women who the cohort, the hazard ratio for gliomas was 0.92 ever took menopausal therapy at age 45 years or (95%CI: 0.54–1.55) in women who ever used above was 0.9 (95%CI: 0.4–2.2), and no trend in menopausal therapy, and no trend in risk with risk with duration of use to 5 years or more was duration of over 3 years of use was observed. observed. In a population-based case–control study of 45 postmenopausal women with gliomas and 2.7 Lymphomas and leukaemias 182 controls in Sweden (Wigertz et al., 2006), the odds ratio in women who ever used meno- Since the previous IARC Monograph IARC pausal therapy was 0.9 (95%CI: 0.4–1.7), and risk (2007), a Danish population-based cohort study did not vary appreciably with duration of use. of menopausal therapy and risk of non-Hodgkin Among 108 cases of meningioma compared to lymphoma was published and was based on 185 controls, the odds ratio for ever use of meno- 157024 women aged 40 years or more of whom pausal therapy was 1.7 (95%CI: 1.0–2.8) but no 23708 were users of menopausal therapy, followed consistent duration–response relationship was for 13 years with linkage to the health service to observed, the risk being highest among those determine menopausal therapy prescriptions who had used menopausal therapy for less than and to the Danish Cancer Registry to identify a year. cases (40 among users and 310 among non-users) In a cohort study based upon records from (Nørgaard et al., 2006). The odds ratio for ever the Mayo clinic of 335318 women aged 26–86 use of menopausal therapy was 0.99 (95%CI: years, 18037 were ever users of menopausal 0.71–1.39), no trend in risk with duration of use therapy (Blitshteyn et al., 2008). Among 1390 to 20 or more years was observed. women with meningioma, 156 were ever users of Ross et al. (2005) evaluated the effect of meno- menopausal therapy. The odds ratio for ever use pausal therapy on risk of leukaemia in a cohort of menopausal therapy was 2.2 (95%CI: 1.9–2.6, of 37172 postmenopausal women (aged 55–69 adjusted for age), with little or no variation by age years at entry) in Iowa, USA. A total of 201 cases over 55 years. No analysis of risk by duration of of leukaemia were identified over 16 years of use was reported. follow-up of which 71 had ever used menopausal therapy. The relative risk for ever use of meno- 2.9 Cancer of the urinary tract pausal therapy was 0.87 (95%CI: 0.65–1.16), with little difference according to type of leukaemia, In a Canadian cohort study of women enrolled and no trend in risk with duration of use of 5 or in a breast cancer screening trial (Kabat et al., more years was observed. 2007), the hazard ratio for renal cancers in women who ever used menopausal therapy was

267 IARC MONOGRAPHS – 100A

0.98 (95%CI: 0.69–1.41), and no trend in risk with In the Canadian cohort study of women duration of use was observed. enrolled in a breast cancer screening trial (Kabat Two cohort studies in the USA showed no et al., 2007), the hazard ratio for cancer in increased risks of cancers of the urinary bladder women who ever used menopausal therapy was in users of menopausal therapy. During approxi- 1.05 (95%CI: 0.85–1.32). No analysis of risk by mately 26 years of follow-up of 116598 women duration of use was reported. enrolled in the Nurse’s Health Study (McGrath Liu et al. (2005) reported on a population- et al., 2006), 22540 were postmenopausal, among based cohort study of 44677 middle-aged never- which 307 cases of bladder cancer were diag- Japanese women, followed for 8–12 nosed. The use of hormones was ascertained years, with 153 lung cancers diagnosed. For periodically during the follow-up period by women with a natural menopause, the relative mailed questionnaire. The relative risk in post- risk from hormone use [not further character- menopausal women with current use of meno- ized] was 1.19 (95%CI: 0.61–2.30), but for women pausal therapy (defined in the study as estrogen with an induced menopause, it was 2.40 (95%CI: plus progestogen) was 0.75 (95%CI: 0.44–1.26, 1.07–5.40). No analysis of risk by duration of use based on 18 cases). No analysis of risk by dura- was reported. [The Working Group considered tion of use was reported. that the majority of women with an induced During an average follow-up of 15.3 years, menopause were likely to have received estrogen 167 cases of bladder cancer developed in a alone.] cohort of 54308 women who were enrolled in the Chen et al. (2007) conducted a case–control Breast Cancer Detection Demonstration Project study of 826 women with lung cancer and 531 (Cantwell. et al., 2006). Menopausal therapy use healthy controls in Taiwan, China. The odds was based on answers to telephone interviews ratio for ever use of menopausal therapy was at the time of recruitment. The relative risk of 0.70 (95%CI: 0.53–0.94). Although no findings bladder cancer was 0.98 (95%CI: 0.71–1.37) in by duration of use were presented, many strati- women who ever used menopausal therapy. No fied analyses were conducted (e.g. by smoking analysis of risk by duration of use was reported. status, age, exposure to cooking fumes, and family history of lung cancer), all showed similar 2.10 Cancer of the lung inverse associations though several of the upper confidence intervals were > 1.0. La Vecchia (2006) reviewed the studies relating to menopausal therapy and lung cancer, 2.11 Cancer of the largely comprising the data considered in the previous IARC Monograph (IARC, 2007), and In a Canadian cohort study of women enrolled concluded that there was no consistent associa- in a breast cancer screening trial (Navarro tion between menopausal therapy and risk for Silvera et al., 2005), the hazard ratio for pancre- lung cancer. atic cancer in women who formerly used meno- A case–control study nested in the Royal pausal therapy was 0.86 (95%CI: 0.55–1.35), and College oral contraceptive study (Elliott & 0.76 (95%CI: 0.47–1.24) among current users. Hannaford, 2006) found no increased risk of lung However, no trend in risk with duration of use cancer in ever users of menopausal therapy. No was observed. analysis of risk by duration of use was reported. The odds ratio for current users (at the time of diagnosis) was 1.2 (95%CI: 0.2–6.4).

268 Combined estrogen-progestogen menopausal therapy

2.12 Cancer of the of combined estrogen–progestogen menopausal therapy increases the risk of cancers of the colon A population-based case–control study in ten or rectum. The Working Group concluded that Canadian provinces (Frise et al., 2006) compared the use of combined estrogen–progestogen answers to a self-administered questionnaire menopausal therapy is unlikely to alter the risk by 326 women with gastric adenocarcinoma to of cancer of the thyroid, lung, stomach, , answers from an equal number of age-matched urinary tract, pancreas, ovary, , or the controls. The odds ratio in women who ever used risk of lymphoma and leukaemia, cutaneous menopausal therapy was 0.72 (95%CI: 0.37–1.40). melanoma, and tumours of the central nervous A slight trend in reduction in risk with increasing system. duration of use was observed, with an odds ratio for 15 or more years of use of 0.42 (95%CI: 0.15–1.16). 3. Cancer in Experimental Animals

2.13 Cancer of the cervix 3.1 Summary of the previous IARC No further evidence has been published that Monograph alters the conclusions reached since the previous of combined hormonal IARC Monograph (IARC, 2007). There is little therapy in mice that are prone to develop evidence to suggest that combined estrogen– mammary tumours resulted in similar inci- progestogen therapy alters the risk for cervical dences of mammary tumours in controls, and in cancer. animals treated with conjugated equine estrogens alone or with conjugated equine estrogens plus 2.14 Cancer of the liver medroxyprogesterone acetate. However, tumour latency was reduced in animals treated with No further evidence has been published that conjugated equine estrogens plus medroxypro- alters the conclusions reached since the previous gesterone acetate. Conjugated equine estrogens IARC Monograph (IARC, 2007). The data for plus medroxyprogesterone acetate suppressed the liver cancer remains too sparse for evaluation. development of uterine (Sakamoto et al., 1997a; IARC, 2007). 2.15 Synthesis Oral administration of conjugated equine estrogens alone or with medroxyprogesterone A large body of evidence was evalu- acetate to ovariectomized rats pretreated with ated for several organ sites, among which the the carcinogen 7,12-dimethylbenz[a]anthracene Working Group concluded combined estrogen– increased the incidence of mammary tumours progestogen menopausal therapy causes cancer with equal frequency, and to a level equal to that of the breast, and of the endometrium. The in non-ovariectomized controls (Sakamoto et al., increased risk for estrogen-induced endome- 1997b; IARC, 2007). trial cancer decreases with the number of days per month that progestogens are added to the regimen. For cancer of the colorectum, the Working Group concluded that it is unlikely that the use

269 IARC MONOGRAPHS – 100A

3.2 Studies published since the Mammary carcinomas were induced with previous IARC Monograph N-methyl-N-nitrosourea at 28 days of age. One group was left untreated (control group), and one 3.2.1 Estradiol and progesterone was subcutaneously implanted with estradiol and (a) Mouse progesterone pellets. Rats that received long- or short-term estradiol plus progesterone treatment Medina et al. (2007) and Rajkumar et al. (2007) had a decreased incidence of any mammary transplanted mammary ducts from the glands carcinomas or of mammary carcinomas with of p53 null BALB/c female mice into the cleared a diameter greater or equal to 1 cm, compared mammary fat pads of p53 wild-type female mice, to control rats. Long-term (but not short-term) and also used the activated Her-2/neu transgenic estradiol plus progesterone treatment increased mouse (FVB) model to show in transplanted the incidence of fibroadenomas. mice and FVB mice that short-term exposure Tsukamoto et al. (2007) used short-term treat- (2 and 3 weeks) to estradiol and progesterone ment with estradiol and progesterone to mimic significantly decreases mammary carcinogenesis in aged female Lewis rats treated in pre-pubertal and mature mice. with N-methyl-N-nitrosourea to show promo- (b) Rat tion of mammary carcinogenesis. Development of N-methyl-N-nitrosourea-induced mammary Blank et al. (2008) used intact and ovariecto- carcinomas was accelerated after short-term mized ACI rats to study the role of progesterone estradiol plus progesterone treatment, compared in mammary carcinogenesis. The animals were with estradiol plus progesterone-untreated rats: subcutaneously implanted with low- or high-dose the incidence of ≥ 1-cm mammary carcinomas estradiol, progesterone alone, low-dose estradiol increased (60 versus 44%), latency was shorter plus progesterone, and ovariectomized ACI rats (28.7 versus 34.6 weeks), and cancer multiplicity with high-dose estradiol plus progesterone. Also, increased significantly (number of all-sized ovariectomized ACI rats were treated with high- carcinomas per rat; 1.8 versus 0.8). dose estradiol plus progesterone plus testosterone The effects of hormones on mammary tumo- propionate to determine the role of the rigenesis were studied by Thordarsonet al. (2004) in hormonal mammary carcinogenesis. In intact in growth-hormone-deficient spontaneous dwarf but not in ovariectomized rats, continuous expo- rats. The rats were divided into several groups sure to high concentrations of estradiol alone treated with: bovine ; estradiol induced mammary carcinogenesis. In ovariec- plus progesterone; bovine growth hormone plus tomized ACI rats, mammary carcinogenesis estradiol plus progesterone; and a control group. require continuous exposure to high concentra- After 1 week, all animals were injected intra- tions of estradiol and progesterone. Testosterone peritoneally with the carcinogen N-methyl-N- had no effect on tumour incidence. nitrosourea. Growth hormone treatment alone 3.2.2 Administration with a known increased mammary tumour incidence from 4.8% in controls to 100%. Estradiol plus proges- carcinogen terone treatment did not significantly alter tumo- (a) Rat rigenesis (0% tumour incidence); estradiol plus Yuri et al. (2006) examined the effects progesterone and growth-hormone obliterated of different durations of exposure to estra- the growth hormone-stimulated increase in diol and progesterone pregnancy levels on tumour development (16.7% tumour incidence). mammary carcinogenesis risk in Lewis rats. See Table 3.1.

270 Combined estrogen-progestogen menopausal therapy null mice BALB/c Comments BALB/c miceBALB/c used in this experiment transplanted were with mammary ducts from p53 < 0.05 < 0.05 < 0.05 < 0.05 < 0.05 < 0.05 < 0.05 Significance P P P P P P P Incidence of Incidence of tumours 15/15 (100%) 2/66 (3%) 3/20 (15%) (20%) 4/20 0/20 (0%) 3/15 (20%) 9/20 (45%) (20%) 4/20 10/20 (50%) 6/15 (37%) Mammary carcinomas (tumours/ transplants): 16/66 (24%) 5/15 (33%)

a a a a

a and P of 15 mg

2 , trans. (3) , trans. (11) and 20 mg P of a a a a 2 /P (5–7) /P (5–7) /P (23–25) /P (5–7) /P (5–7) 3 wk) /P (for (for 3 wk) (for 2 2 2 2 2 2 2 Animals/group at start at Animals/group Route regimen Dosing Initial number: NR Untreated E s.c. implant 100 μg E of E E Untreated E Untreated, trans. (3) Untreated, trans. (11) E E Initial number: NR Untreated E s.c. implant 50 µg E of

. . et al

et al

Reference Duration 45–58 wk Rajkumar Table 3.1 Studies of cancer in experimental 3.1 animalsTable exposed to estrogen–progestogen combinations Species, strain (sex) Mouse, BALB/c (F) 24–32 wk Mouse, FVB (F) (2007) Medina (2007)

271 IARC MONOGRAPHS – 100A < 0.01) P /P group developed fibroadenomas 0/20vs 2 /P pellet was replaced every 3–4 wk (long-term) or 2 Comments The E MNU-treated rats ( implanted only once (short-term); 5/14 rats The of implantedMNU 5/14 only once (short-term); + long-term E /P vs 2

< 0.01, < 0.01, < 0.01 < 0.007 < 0.0001 < 0.0001 < 0.01, < 0.01, < 0.01 NS NS NS Significance P P MNU + E MNU NS NS NS P NS P P P P 0/10 0/15 0/15 0/14 Incidence of Incidence of tumours 4/19 11/19 (21%), (58%) 0/15 0/8 0/15 Mammary carcinomas ≥ 1 cm in diameter, any carcinoma: 19/20 (95%), 20/20 20/20 (95%), 19/20 (100%) 5/7 0/15 11/11 14/14 5/14 (36%), 9/14 (64%) Mammary carcinomas: 0/10

/P /P 2 2

(30→3 mg) plus 30 mg P of plus 30 mg P plus of 30 mg 2 2 2 2 (10 μg) plus 30 mg P of plus 30 mg P of 2 2 2 2 /P s.c. implant: 0.5 mg/32.5 mg 2 Low E High E s.c. implant – ovariectomized rats Untreated control High E Animals/group at start at Animals/group Route regimen Dosing High E 30 mg P Low E Initial number: NR MNU MNU + Long-term E Low E of TP 30 mg P High E MNU + Short-term E s.c. implant – intact rats Initial number: NR Untreated control Low E MNU: 50 mg/kg i.p. bw E

. (2008)

. (2006) et al

et al Reference Duration Table 3.1 (continued) 3.1 Table Species, strain (sex) Rat, (F) ACI 39 wk Rat, Lewis (F) 6 wk Blank on Dependent tumour volume wk 29 or (≥ 1 cm) Yuri

272 Combined estrogen-progestogen menopausal therapy Comments Spontaneous dwarf rats are growth-hormone-deficient + 2 < 0.003] < 0.05 P P vs bGH) Significance P [ (bGH and E

-nitrosourea; mo, month or months; NR, not reported; NS, not significant;progesterone;P, N

-methyl-

< 0.05) Incidence of Incidence of tumours 1/21 (4.8%) 8/8 (100%) Mammary carcinomas: Mammary carcinomas: vs 9/15 (44%) 11/25 (60%) multiplicity:Tumour 0.8 ± 0.2 vs 1.8 ± 0.5 Latency was shorter vs(28.7 34.6 wk, P 1/6 (16.7%) 0/11 (0%)

, estradiol; female; F, MNU, N 2

+ P 2 (0.5 mg)/P (32.5 mg), one (0.5 mg)/P one (32.5 mg), 2 (30 µg) + P (30 mg/2 mo) (s.c.) (30 µg) + P (30 mg/2 mo) 2 Animals/group at start at Animals/group Control (MNU only) bGH E Route regimen Dosing MNU, 20 mg/kg 7 wk at of (i.p.) bw age MNU, 20 mg/kg 7 wk at of (i.p.) bw age + E bGH, 40–50 mg/kg in bw 50 µl weekly s.c. injections; MNU, 50 mg/ 1 wk afterhormone kg (i.p.), bw treatment pellet s.c. implanted 24 at wk 29 animals/group Initial number NR bGH and E . .

et al et al

from week x to week y or at week x

Reference Duration Table 3.1 (continued) 3.1 Table Species, strain (sex) Rat, Lewis (F) Rat, (F) SDR 48 wk Tsukamoto (2007) 20 wk Thordarson (2004) TP, testosteroneTP, propionate; trans., transplanted; i.p., intraperitoneal; s.c., subcutaneously; vs, versus; wk, week or weeks a bGH, bovine growth hormone; body bw, weight; E

273 IARC MONOGRAPHS – 100A

4. Other Relevant Data (Boyapati et al., 2005). In a case–control study of American women, sequence variations in the 4.1 Absorption, distribution, of the and eight involved in the of estrogen metabolism, and were assessed in relation to endometrial cancer Various combinations of estrogens and risk. Women with a particular polymorphism in progestogens are used for hormonal menopausal the SULT1A1 gene were found to have a signifi- therapy. Because penetrate normal skin cantly elevated risk of endometrial cancer if they easily, a variety of systems have been developed received HRT (Rebbeck et al., 2006). In another that deliver estrogens and progestogens paren- study of American women with breast cancer terally (e.g. transdermal patches), thus bypassing and their case–controls, polymorphisms in the the liver. genes for progesterone receptor and CYP3A4 While the mechanisms of absorption and were assessed in relation to breast cancer. It was distribution of estrogens and progestogens have reported that there was an elevated risk of ductal been known for several years, only recently breast cancer or PR+ breast cancers among has an understanding of the genes that encode women with PGR331A alleles and greater than the enzymes which control the enzymatic 3 years of combined HRT use. Women with at steps involved in metabolism been least one CYP3A4*1B allele who did not have a acquired. This applies especially to the oxidative history of HRT use had a higher risk of ER- breast metabolism of estrogens. The phase I enzymes cancers (Rebbeck et al., 2007). 1A1 and 1B1 catalyse the production of catechol estrogens further oxidized 4.2 Genetic and related effects to estrogen quinones that can induce the forma- tion of DNA adducts. This is counteracted by the 4.2.1 Direct genotoxicity phase II enzymes, catechol-O-methyltransferase Data on the genetic effects of estrogens and and glutathione S-transferase P1, which reduce their derivatives indicate that these compounds the levels of catechol and quinones by forming give rise to reactive and reactive methoxyestrogens and glutathione conju- oxygen species that can induce DNA damage. In gates. Polymorphic variants of these and other recent years, it has been reported that metabo- enzymes occur frequently in the population and lites of estrogen can form adducts on DNA and, several are associated with altered func- based on this, it has been suggested that these tion (IARC, 2007). lesions could induce the genetic alterations found One of the particular areas of research on in cancers (IARC, 2007; see also Monograph genetic variations concerns the capacity to metab- on Estrogen-only Menopausal Therapy in olize hormones. Two common polymorphisms this volume). The evidence reported since the in the cytochrome P450 1A1 (CYP1A1) gene were previous evaluation further substantiates the examined in Chinese women with or without premise that these mechanisms could contribute breast cancer. Homozygosity for both alleles was to the induction of cancer by estrogens. Since associated with a reduction of risk of borderline the previous IARC Monograph (IARC, 2007), significance and the reduction was greater in it has been shown that DNA adducts derived slender women or those with a long history of from equine estrogens can interfere with DNA menstrual cycles. Use of hormone replacement synthesis. Using an in-vitro DNA primer exten- therapy (HRT) was not considered in this study sion assay past adducts with bypass polymerases

274 Combined estrogen-progestogen menopausal therapy kappa and eta, it has been shown that 4-hydrox- findings have been made in individual studies yequilenin adducts are highly mutagenic giving but their interpretation awaits repetition of find- rise to both A→T transversions and A→G tran- ings particularly in other laboratories and other sitions. This finding offers a plausible mecha- study populations.] nism for the contribution of estrogen adducts Other studies have considered the relation- to the induction of cancer (Yasui et al., 2006). ships between various genetic variants and the [The Working Group noted that although these risk of colon cancer. In a study considering new findings increase the plausibility of these polymorphisms in the genes for IGF-1 and the pathways as mechanisms of estrogen-related IGF-binding 3 (IGF-BP3), women with carcinogenesis, they do not prove that these are the GG of IGF-BP3 who received the major pathways to estrogen-related cancers. HRT had a reduced incidence of colon cancer The way in which progestogens might influence compared with women who did not receive HRT the genotoxicity of estrogens is not known.] (Morimoto et al., 2005). Genetic variants of ERα, Targeted studies have explored how sequence ERβ, and the were studied in variations in specific genes influence cancer risk, relationship to colorectal cancer. It was found that including their effects on hormone receptor women having an R allele (absence of an Rsa1 function or metabolism of hormones. One of restriction site) at 1,082G > A in ERβ had reduced these recently published studies showed that risk of colorectal cancer if they received HRT certain genetic variants in the ERα gene or (Slattery et al., 2005). In another study, SNPs were progesterone receptor genes were linked to evaluated in ten estrogen-metabolism-related increased mammographic density following genes. None of the studied SNPs were associ- HRT (van Duijnhoven et al., 2006). In another ated with an elevated risk of colon cancer. There study, a specific mutation in the ERα was shown were no significant differences between women to be linked to a higher risk of breast cancer receiving HRT and those who did not (Huber (Conway et al., 2007). In contrast, in a study of et al., 2005). In a study reflecting the expression Hispanic and non-Hispanic women in the South levels of DNA mismatch repair genes and meas- western USA, particular genetic variants in the uring the frequencies of mismatch repair defects ERα or androgen receptor genes were not found detected in single- and di- repeats, to be related to breast cancer risk (Slattery et al., use of estrogen or estrogen–progestogen combi- 2007a). In another study by the same group and nations were evaluated for their effect on colo- the same study population, genetic variants in rectal cancer incidences. For all women taking the genes of the -related pathways were combinations of estrogen and progestogen, there assessed with regard to breast cancer risk. In was a reduced risk of colorectal cancer. For those this study, some of the variants examined were women with little or no evidence of mismatch related to breast cancer and to an involvement repair defects, the colon cancer risk reduction with HRT, with positive or negative relation- was greater (40%). Neither of these effects was ships observed for particular variants (Slattery observed in women treated with estrogen alone et al., 2007b). [The Working Group noted that (Newcomb et al., 2007). the study of single-nucleotide polymorphisms (SNPs) and other sequence variations as factors affecting the risk of breast cancer and increased breast density in patients treated with estrogen plus progestogen menopausal therapy are at an early stage of development. Some intriguing

275 IARC MONOGRAPHS – 100A

4.2.2 Receptor-mediated effects for estradiol and various equine estrogens with unsaturated B-rings found in conjugated equine The literature on the receptor-mediated effects estrogens used for HRT. Differences in binding of estrogen–progestogen menopausal therapy of these estrogens to the -binding domain was reviewed in the previous IARC Monograph of ERα were determined by crystallography. In (IARC, 2007). comparison to binding by estradiol, decreased In more recently reported studies, colon ligand flexibility and hydrophobicity was found cancer cells were transfected to transiently for the equine estrogens with unsaturated B-rings overexpress ERα and cultured with or without (Hsieh et al., 2008). estrogen. Gene expression for hTNF-α and DNA The effects of synthetic progestogens fragmentation were evaluated as measures of (progestins) on androgen receptors were consid- and β-catenin signalling was evalu- ered in a review article. Evidence was reviewed ated as a measure of cell proliferation. ERα over- indicating that synthetic hormones that act like expression, with or without estrogen treatment, progestins may exert their effect in part through activated DNA fragmentation. ERα overexpres- their ability to bind to androgen receptors sion combined with estrogen treatment increased and, through linked pathways, act to suppress both expression of hTNF-α and DNA fragmen- estrogen-induced functions. The activity of these tation, and treatment of cells with antibodies progestins occurs independently of their effects to hTNF-α reduced the DNA fragmentation. through the progesterone receptors. The authors ERα overexpression combined with estrogen propose that some of the reported excess of breast treatment increased expression of proliferation cancer associated with synthetic progestins, such inhibitory molecules p21 and p27 and decreased as medroxyprogesterone acetate, may occur expression of β-catenin and its downstream pro- because of their endocrine receptor effects on the proliferation target genes, cyclin D1 and Rb (Hsu androgen receptor (Birrell et al., 2007). et al., 2006). [The Working Group noted that this The relationship between radiological breast would be consistent with an estrogen-mediated density and breast histology was evaluated in a inhibition of cell proliferation in the colon.] study of American breast cancer patients treated The expression of steroid receptor coacti- with HRT as compared to an equal number of vator AIB1 was examined in endometrial speci- patients not being treated with HRT. Studies mens from patients with endometrial cancer, focused on areas of the breast not involved with comparing areas with cancer to normal areas cancer and evaluated radiological breast density or to those with complex atypical . and histological fibrous stroma, ducts and Expression of AIB1 was assessed by immu- lobules, and evidence of cell proliferation rates. nohistochemistry in comparison to ERα and The higher breast density in patients receiving progesterone receptors. AIB1 was most highly HRT was correlated significantly with the pres- expressed in endometrial cancers. There were no ence of fibrous stroma and type I lobules, as well differences detected between the morphological as increased proliferation in ducts and lobules. groups for the expression of other co-regulators and progesterone receptor tested. It was suggested that when AIB1 and ER levels in the breast did not correlate with are expressed together, ER activity is enhanced, HRT therapy or breast density. Because the contributing to hyperplasia and malignancy increase in breast stroma and type I lobules (Balmer, et al., 2006). was not related to hormone receptor levels, it is Transcriptional activation in vivo by ERα, speculated that this effect may be mediated by ERβ and the androgen receptor were compared paracrine factors (Harvey et al., 2008).

276 Combined estrogen-progestogen menopausal therapy

In an in-vitro study, medium conditioned carcinogenesis by combined estrogen progestogen by progesterone-treated human breast cancer menopausal therapy . There is also support for cells was shown to produce paracrine factors the potential involvement of genotoxic effects that induced the proliferation of endothelial of combined estrogen–progestogen menopausal cells and epithelial breast tumour cells through therapy estrogenic hormones or their associated VEGF receptors. Inhibition of VEGF receptor 1 metabolic by-products including the formation by antibody and VEGF receptor 2 using SU-1498 of DNA adducts, and reactive oxygen species that blocked the induced proliferation of the epithe- damage DNA. Recent data suggests that these lial breast tumour cells and endothelial cells, adducts slow down or block DNA replication as did anti-progestin (RU486) and invoke bypass replication, which is prone to treatment of the tumour cells that conditioned mutagenesis. the medium. These results were interpreted as The predominant effects of combined implicating VEGF and possibly other paracrine estrogen–progestogen menopausal therapy asso- factors in the progestin-induced proliferation of ciated with hormonal carcinogenesis are likely to endothelial cells surrounding breast tumours be the result of one or more receptor-mediated (Liang & Hyder, 2005). processes. Progestogens including those used In a study of ER+ and ER- breast cell lines, for combined estrogen–progestogen menopausal both malignant and non-malignant, it was shown therapy appear to have the capacity to stimu- that both estradiol and iron (as ferrous sulfate) late cell proliferation in the breast while they increased cell proliferation as measured by Ki-67 inhibit proliferation in the uterus. The magni- and proliferating cell nuclear antigen levels and tude of these effects vary for different synthetic that the combination of estradiol and iron caused progestogens, with a suggestion that medroxy- even greater increases (Dai et al., 2008). progesterone acetate is very active. Norwegian women on systemic HRT had Cessation of hormonal treatment may reduce elevated plasma estrogen levels that were compa- some receptor-mediated effects. The hormone- rable to those in premenopausal women. Among induced genotoxic effects may be persistent. women not on HRT, plasma levels of estradiol Use of combined estrogen–progestogen and serum hormone-binding globulin were menopausal therapy was linked to increases in influenced by the women’s basal metabolic index breast density, which is an established risk factor (Waaseth et al., 2008). In a study of Swedish ER+ for breast cancer. breast cancer patients, HRT caused altered expres- sion of 276 genes as compared to their expres- sion in a larger control group of ER+ tumours 5. Evaluation from patients with unknown HRT usage. It was concluded that patients with postmenopausal There is sufficient evidence in humans for HRT use had lower ER protein levels, a distinct the carcinogenicity of combined estrogen– gene expression profile, and better disease-free progestogen menopausal therapy. Combined survival (Hall et al., 2006). estrogen–progestogen menopausal therapy causes cancer of the breast, and of the endome- 4.3 Synthesis trium. The increased risk for estrogen-induced endometrial cancer decreases with the number Current knowledge indicates that hormone- of days per month that progestogens are added receptor-mediated responses are a plausible and to the regimen. probably necessary mechanism for hormonal

277 IARC MONOGRAPHS – 100A

There is limited evidence in experimental may increase risk of developing breast cancerFASEB J, 21: animals for the carcinogenicity of conjugated 2285–2293. doi:10.1096/fj.06-7518com PMID:17413000 Blank EW, Wong PY, Lakshmanaswamy R et al. (2008). equine estrogens plus medroxyprogesterone Both ovarian hormones estrogen and progesterone acetate. are necessary for hormonal mammary carcinogen- There is limited evidence in experimental esis in ovariectomized ACI ratsProc Natl Acad Sci U S A, 105: 3527–3532. doi:10.1073/pnas.0710535105 animals for the carcinogenicity of estradiol plus PMID:18299580 progesterone. Blitshteyn S, Crook JE, Jaeckle KA (2008). Is there an asso- Combined estrogen–progestogen meno- ciation between meningioma and hormone replace- pausal therapy is carcinogenic to humans ment therapyJ Clin Oncol, 26: 279–282. doi:10.1200/ JCO.2007.14.2133 PMID:18182668 (Group 1). Bouchardy C, Morabia A, Verkooijen HM et al. (2006). Remarkable change in age-specific breast cancer inci- dence in the Swiss canton of Geneva and its possible References relation with the use of hormone replacement thera- pyBMC Cancer, 6: 78 doi:10.1186/1471-2407-6-78 PMID:16551373 Allemand H, Seradour B, Weill A, Ricordeau P (2008). Boyapati SM, Shu XO, Gao YT e t al . (2005). Polymorphisms [Decline in breast cancer incidence in 2005 and 2006 in CYP1A1 and breast carcinoma risk in a population- in France: a paradoxical trend]Bull Cancer, 95: 11–15. based case-control study of Chinese womenCancer, 103: PMID:18230565 2228–2235. doi:10.1002/cncr.21056 PMID:15856430 Anderson GL, Chlebowski RT, Rossouw JE et al. (2006). Campagnoli C, Clavel-Chapelon F, Kaaks R et al. (2005). Prior hormone therapy and breast cancer risk in the Progestins and progesterone in hormone replacement Women’s Health Initiative randomized trial of estrogen therapy and the risk of breast cancerJ Steroid Biochem plus progestinMaturitas, 55: 103–115. doi:10.1016/j. Mol Biol, 96: 95–108. doi:10.1016/j.jsbmb.2005.02.014 maturitas.2006.05.004 PMID:16815651 PMID:15908197 Anderson GL, Judd HL, Kaunitz AM et al.Women’s Health Cantwell MM, Lacey JV Jr, Schairer C et al. (2006). Initiative Investigators. (2003). Effects of estrogen Reproductive factors, exogenous hormone use and plus progestin on gynecologic cancers and associated bladder cancer risk in a prospective studyInt J Cancer, diagnostic procedures: the Women’s Health Initiative 119: 2398–2401. doi:10.1002/ijc.22175 PMID:16894568 randomized trialJAMA, 290: 1739–1748. doi:10.1001/ Chen KY, Hsiao CF, Chang GC et al.GEFLAC Study jama.290.13.1739 PMID:14519708 Group. (2007). Hormone replacement therapy and Balmer NN, Richer JK, Spoelstra NS et al. (2006). Steroid lung cancer risk in ChineseCancer, 110: 1768–1775. receptor coactivator AIB1 in endometrial carcinoma, doi:10.1002/cncr.22987 PMID:17879370 hyperplasia and normal endometrium: Correlation Clarke CA, Glaser SL, Uratsu CS et al. (2006). Recent with clinicopathologic parameters and biomarkersMod declines in hormone therapy utilization and breast Pathol, 19: 1593–1605. doi:10.1038/modpathol.3800696 cancer incidence: clinical and population-based PMID:16980945 evidenceJ Clin Oncol, 24: e49–e50. doi:10.1200/ Beral V, Banks E, Reeves G, Bull DMillion Women JCO.2006.08.6504 PMID:17114650 Study Collaborators. (2003). Breast cancer and Col NF, Kim JA, Chlebowski RT (2005). Menopausal hormone-replacement therapy in the Million Women hormone therapy after breast cancer: a meta-analysis StudyLancet, 362: 1330–1331. doi:10.1016/S0140- and critical appraisal of the evidenceBreast Cancer Res, 6736(03)14596-5 PMID:12927427 7: R535–R540. doi:10.1186/bcr1035 PMID:15987460 Beral V, Bull D, Green J, Reeves GMillion Women Collins JA, Blake JM, Crosignani PG (2005). Breast cancer Study Collaborators. (2007). Ovarian cancer and risk with postmenopausal hormonal treatmentHum hormone replacement therapy in the Million Women Reprod Update, 11: 545–560. doi:10.1093/humupd/ StudyLancet, 369: 1703–1710. doi:10.1016/S0140- dmi028 PMID:16150813 6736(07)60534-0 PMID:17512855 Conway K, Parrish E, Edmiston SN et al. (2007). Risk Beral V, Bull D, Reeves GMillion Women Study factors for breast cancer characterized by the estrogen Collaborators. (2005). Endometrial cancer and receptor alpha A908G (K303R) mutationBreast Cancer hormone-replacement therapy in the Million Women Res, 9: R36 doi:10.1186/bcr1731 PMID:17553133 StudyLancet, 365: 1543–1551. doi:10.1016/S0140- Corrao G, Zambon A, Conti V et al. (2008). Menopause 6736(05)66455-0 PMID:15866308 hormone replacement therapy and cancer risk: an Italian Birrell SN, Butler LM, Harris JM et al. (2007). Disruption record linkage investigationAnn Oncol, 19: 150–155. of androgen receptor signaling by synthetic progestins doi:10.1093/annonc/mdm404 PMID:17785762

278 Combined estrogen-progestogen menopausal therapy

Dai J, Jian J, Bosland M et al. (2008). Roles of hormone therapy use: correlation with breast density, estrogen replacement therapy and iron in proliferation of breast receptor, progesterone receptor, and proliferation indi- epithelial cells with different estrogen and proges- cesMenopause, 15: 67–73. PMID:17558338 terone receptor statusBreast, 17: 172–179. doi:10.1016/j. Hausauer AK, Keegan TH, Chang ET, Clarke CA (2007). breast.2007.08.009 PMID:17928227 Recent breast cancer trends among Asian/Pacific Danforth KN, Tworoger SS, Hecht JL et al. (2007). A Islander, Hispanic, and African-American women in prospective study of postmenopausal hormone use the US: changes by tumor subtypeBreast Cancer Res, 9: and ovarian cancer riskBr J Cancer, 96: 151–156. R90 doi:10.1186/bcr1839 PMID:18162138 doi:10.1038/sj.bjc.6603527 PMID:17179984 Heiss G, Wallace R, Anderson GL et al.WHI Investigators. Dinger JC, Heinemann LA, Möhner S et al. (2007). Colon (2008). Health risks and benefits 3 years after stopping cancer risk and different HRT formulations: a case- randomized treatment with estrogen and progestin- control studyBMC Cancer, 7: 76 doi:10.1186/1471-2407- JAMA, 299: 1036–1045. doi:10.1001/jama.299.9.1036 7-76 PMID:17488513 PMID:18319414 Dumeaux V, Fournier A, Lund E, Clavel-Chapelon F Holmberg L & Anderson HHABITS steering and data (2005). Previous oral contraceptive use and breast monitoring committees. (2004). HABITS (hormonal cancer risk according to hormone replacement therapy replacement therapy after breast cancer–is it safe?), use among postmenopausal womenCancer Causes a randomised comparison: trial stoppedLancet, Control, 16: 537–544. doi:10.1007/s10552-004-8024-z 363: 453–455. doi:10.1016/S0140-6736(04)15493-7 PMID:15986108 PMID:14962527 Elliott AM & Hannaford PC (2006). Use of exogenous Holmberg L, Iversen OE, Rudenstam CM et al.HABITS hormones by women and lung cancer: evidence Study Group. (2008). Increased risk of recurrence after from the Royal College of General Practitioners’ hormone replacement therapy in breast cancer survi- Oral Contraception StudyContraception, 73: vorsJ Natl Cancer Inst, 100: 475–482. doi:10.1093/jnci/ 331–335. doi:10.1016/j.contraception.2005.10.003 djn058 PMID:18364505 PMID:16531161 Hsieh RW, Rajan SS, Sharma SK, Greene GL (2008). Fournier A, Berrino F, Riboli E et al. (2005). Breast cancer Molecular characterization of a B-ring unsatu- risk in relation to different types of hormone replace- rated estrogen: implications for conjugated equine ment therapy in the E3N-EPIC cohortInt J Cancer, 114: estrogen components of premarinSteroids, 73: 59–68. 448–454. doi:10.1002/ijc.20710 PMID:15551359 doi:10.1016/j.steroids.2007.08.014 PMID:17949766 Frise S, Kreiger N, Gallinger S et al. (2006). Menstrual and Hsu HH, Cheng SF, Chen LM et al. (2006). Over-expressed reproductive risk factors and risk for gastric adenocar- estrogen receptor-alpha up-regulates hTNF-alpha cinoma in women: findings from the canadian national gene expression and down-regulates beta-catenin enhanced cancer surveillance systemAnn Epidemiol, signaling activity to induce the apoptosis and inhibit 16: 908–916. doi:10.1016/j.annepidem.2006.03.001 proliferation of LoVo colon cancer cellsMol Cell PMID:16843679 Biochem, 289: 101–109. doi:10.1007/s11010-006-9153-3 Glass AG, Lacey JV Jr, Carreon JD, Hoover RN (2007). PMID:16628468 Breast cancer incidence, 1980–2006: combined roles Huber A, Bentz EK, Schneeberger C et al. (2005). Ten of menopausal hormone therapy, screening mammog- polymorphisms of estrogen-metabolizing genes raphy, and estrogen receptor statusJ Natl Cancer Inst, and a family history of colon cancer–an association 99: 1152–1161. doi:10.1093/jnci/djm059 PMID:17652280 study of multiple gene-gene interactionsJ Soc Gynecol Greiser CM, Greiser EM, Dören M (2005). Menopausal Investig, 12: e51–e54. doi:10.1016/j.jsgi.2005.07.003 hormone therapy and risk of breast cancer: a meta- PMID:16202920 analysis of epidemiological studies and randomized IARC (1999). and Post- controlled trialsHum Reprod Update, 11: 561–573. Menopausal Hormonal TherapyIARC Monogr Eval doi:10.1093/humupd/dmi031 PMID:16150812 Carcinog Risks Hum, 72: 1–660. Greiser CM, Greiser EM, Dören M (2007). Menopausal IARC (2007). Combined estrogen–progestogen contra- hormone therapy and risk of ovarian cancer: ceptives and combined estrogen–progestogen meno- systematic review and meta-analysisHum Reprod pausal therapyIARC Monogr Eval Carcinog Risks Hum, Update, 13: 453–463. doi:10.1093/humupd/dmm012 91: 1–528. PMID:18756632 PMID:17573406 Jemal A, Ward E, Thun MJ (2007). Recent trends in breast Hall P, Ploner A, Bjöhle J et al. (2006). Hormone- cancer incidence rates by age and tumor character- replacement therapy influences gene expression istics among U.S. womenBreast Cancer Res, 9: R28 profiles and is associated with breast-cancer prognosis: doi:10.1186/bcr1672 PMID:17477859 a cohort studyBMC Med, 4: 16 PMID:16813654 Kabat GC, Miller AB, Rohan TE (2007). Reproductive and Harvey JA, Santen RJ, Petroni GR et al. (2008). Histologic hormonal factors and risk of lung cancer in women: a changes in the breast with menopausal hormone

279 IARC MONOGRAPHS – 100A

prospective cohort studyInt J Cancer, 120: 2214–2220. Liu Y, Inoue M, Sobue T, Tsugane S (2005). Reproductive doi:10.1002/ijc.22543 PMID:17278095 factors, hormone use and the risk of lung cancer among Katalinic A & Rawal R (2008). Decline in breast cancer inci- middle-aged never-smoking Japanese women: a large- dence after decrease in utilisation of hormone replace- scale population-based cohort studyInt J Cancer, 117: ment therapyBreast Cancer Res Treat, 107: 427–430. 662–666. doi:10.1002/ijc.21229 PMID:15929081 doi:10.1007/s10549-007-9566-z PMID:17453336 Lund E, Bakken K, Dumeaux V et al. (2007). Hormone Kerlikowske K, Miglioretti DL, Buist DS et al.National replacement therapy and breast cancer in former users Cancer Institute-Sponsored Breast Cancer Surveillance of oral contraceptives–The Norwegian Women and Consortium. (2007). Declines in invasive breast cancer Cancer studyInt J Cancer, 121: 645–648. doi:10.1002/ and use of postmenopausal hormone therapy in a ijc.22699 PMID:17372914 screening mammography populationJ Natl Cancer Inst, McEvoy GK, editor (2007) 2007 AHFS Drug Information, 99: 1335–1339. doi:10.1093/jnci/djm111 PMID:17698950 Bethesda, MD, American Society of Health-System Krieger N (2008). Hormone therapy and the rise and Pharmacists, American Hospital Formulary Service, perhaps fall of US breast cancer incidence rates: critical pp. 1186–1194. reflectionsInt J Epidemiol, 37: 627–637. doi:10.1093/ije/ McGrath M, Michaud DS, De Vivo I (2006). Hormonal dyn055 PMID:18375445 and reproductive factors and the risk of bladder cancer La Vecchia C (2006). Hormone replacement therapy in in womenAm J Epidemiol, 163: 236–244. doi:10.1093/ menopause and lung cancer: an updateEur J Cancer aje/kwj028 PMID:16319290 Prev, 15: 283–284. doi:10.1097/00008469-200608000- Medina D, Kittrell FS, Tsimelzon A, Fuqua SA (2007). 00001 PMID:16835498 Inhibition of mammary tumorigenesis by estrogen Lacey JV Jr, Brinton LA, Leitzmann MF et al. (2006). and progesterone in genetically engineered mice. Ernst Menopausal hormone therapy and ovarian cancer risk Schering Found Symp Proc, 1:109–126. in the National Institutes of Health-AARP Diet and Moorman PG, Schildkraut JM, Calingaert B et al. Health Study CohortJ Natl Cancer Inst, 98: 1397–1405. (2005). Menopausal hormones and risk of ovarian doi:10.1093/jnci/djj375 PMID:17018786 cancer. Am J Obstet Gynecol, 193: 76–82. doi:10.1016/j. Lacey JV Jr, Brinton LA, Lubin JH e t al . (2005). Endometrial ajog.2004.11.013 PMID:16021062 carcinoma risks among menopausal estrogen plus Morimoto LM, Newcomb PA, White E et al. (2005). progestin and unopposed estrogen users in a cohort of Insulin-like growth factor polymorphisms and colo- postmenopausal womenCancer Epidemiol rectal cancer risk. Cancer Epidemiol Biomarkers Prev, Prev, 14: 1724–1731. doi:10.1158/1055-9965.EPI-05- 14: 1204–1211. doi:10.1158/1055-9965.EPI-04-0695 0111 PMID:16030108 PMID:15894673 Lacey JV Jr, Leitzmann MF, Chang SC et al. (2007). Naldi L, Altieri A, Imberti GL et al.Oncology Study Group Endometrial cancer and menopausal hormone of the Italian Group for Epidemiologic Research in therapy in the National Institutes of Health-AARP Dermatology (GISED). (2005). Cutaneous malignant Diet and Health Study cohortCancer, 109: 1303–1311. melanoma in women. Phenotypic characteristics, sun doi:10.1002/cncr.22525 PMID:17315161 exposure, and hormonal factors: a case-control study Lea CS, Holly EA, Hartge P et al. (2007). Reproductive from Italy. Ann Epidemiol, 15: 545–550. doi:10.1016/j. risk factors for cutaneous melanoma in women: a case- annepidem.2004.10.005 PMID:16029848 control studyAm J Epidemiol, 165: 505–513. doi:10.1093/ Navarro Silvera SA, Miller AB, Rohan TE (2005). aje/kwk040 PMID:17158470 Hormonal and reproductive factors and pancreatic Lee S, Kolonel L, Wilkens L et al. (2006). Postmenopausal cancer risk : a . Pancreas, 30: hormone therapy and breast cancer risk: the Multiethnic 363–374. doi:10.1097/01.mpa.0000160301.59319.ba CohortInt J Cancer, 118: 1285–1291. doi:10.1002/ PMID:15841049 ijc.21481 PMID:16170777 Nazeri K, Khatibi A, Nyberg P et al. (2006). Colorectal Li CI, Daling JR, Malone KE et al. (2006). Relationship cancer in middle-aged women in relation to hormonal between established breast cancer risk factors and risk status: a report from the Women’s Health in the Lund of seven different histologic types of invasive breast Area (WHILA) study. Gynecol Endocrinol, 22: 416–422. cancerCancer Epidemiol Biomarkers Prev, 15: 946–954. doi:10.1080/09513590600900378 PMID:17012102 doi:10.1158/1055-9965.EPI-05-0881 PMID:16702375 Newcomb PA, Zheng Y, Chia VM et al. (2007). Estrogen Liang Y & Hyder SM (2005). Proliferation of endothe- plus progestin use, microsatellite instability, and the lial and tumor epithelial cells by progestin-induced risk of colorectal cancer in women. Cancer Res, 67: vascular endothelial growth factor from human breast 7534–7539. doi:10.1158/0008-5472.CAN-06-4275 cancer cells: paracrine and autocrine effectsEndo- PMID:17671225 crinology, 146: 3632–3641. doi:10.1210/en.2005-0103 Nørgaard M, Poulsen AH, Pedersen L et al. (2006). Use PMID:15845615 of postmenopausal hormone replacement therapy and risk of non-Hodgkin’s lymphoma: a Danish

280 Combined estrogen-progestogen menopausal therapy

population-based cohort study. Br J Cancer, 94: 1339– epithelial ovarian cancer. Cancer Epidemiol Biomarkers 1341. doi:10.1038/sj.bjc.6603123 PMID:16670705 Prev, 16: 2548–2556. doi:10.1158/1055-9965.EPI-07- Prentice RL, Chlebowski RT, Stefanick ML et al. (2008). 0550 PMID:18086757 Estrogen plus progestin therapy and breast cancer in Rossouw JE, Anderson GL, Prentice RL et al.Writing recently postmenopausal women. Am J Epidemiol, 167: Group for the Women’s Health Initiative Investigators. 1207–1216. doi:10.1093/aje/kwn044 PMID:18372396 (2002). Risks and benefits of estrogen plus progestin in Rajkumar L, Kittrell FS, Guzman RC et al. (2007). healthy postmenopausal women: principal results From Hormone-induced protection of mammary tumori- the Women’s Health Initiative randomized controlled genesis in genetically engineered mouse models. Breast trial. JAMA, 288: 321–333. doi:10.1001/jama.288.3.321 Cancer Res, 9: R12 doi:10.1186/bcr1645 PMID:17257424 PMID:12117397 Ravdin PM, Cronin KA, Howlader N et al. (2007). The Sakamoto S, Kudo H, Suzuki S et al. (1997b). Additional decrease in breast-cancer incidence in 2003 in the effects of medroxyprogesterone acetate on mammary United States. N Engl J Med, 356: 1670–1674. doi:10.1056/ tumors in oophorectomized, estrogenized, DMBA- NEJMsr070105 PMID:17442911 treated rats. Anticancer Res, 17: 6D4583–4587. Rebbeck TR, Troxel AB, Norman S et al. (2007). PMID:9494572 Pharmacogenetic modulation of combined hormone Sakamoto S, Mori T, Shinoda H et al. (1997a). Effects of replacement therapy by progesterone-metabolism conjugated estrogens with or without medroxyproges- genotypes in postmenopausal breast cancer risk. Am terone acetate on mammary carcinogenesis, uterine J Epidemiol, 166: 1392–1399. doi:10.1093/aje/kwm239 adenomyosis and femur in mice. Acta Anat (Basel), 159: PMID:17827444 204–208. doi:10.1159/000147985 PMID:9605604 Rebbeck TR, Troxel AB, Wang Y et al. (2006). Estrogen Shah NR, Borenstein J, Dubois RW (2005). genes, hormone replacement therapy, and Postmenopausal hormone therapy and breast cancer: endometrial cancer risk. J Natl Cancer Inst, 98: 1311– a systematic review and meta-analysis. Menopause, 1320. doi:10.1093/jnci/djj360 PMID:16985250 12: 668–678. doi:10.1097/01.gme.0000184221.63459.e1 Reeves GK, Beral V, Green J et al.Million Women Study PMID:16278609 Collaborators. (2006). Hormonal therapy for meno- Silvera SA, Miller AB, Rohan TE (2006). Hormonal and pause and breast-cancer risk by histological type: reproductive factors and risk of glioma: a prospective a cohort study and meta-analysis. Lancet Oncol, cohort study. Int J Cancer, 118: 1321–1324. doi:10.1002/ 7: 910–918. doi:10.1016/S1470-2045(06)70911-1 ijc.21467 PMID:16152609 PMID:17081916 Slattery ML, Sweeney C, Herrick J et al. (2007a). ESR1, Robbins AS & Clarke CA (2007). Regional changes in AR, body size, and breast cancer risk in Hispanic and hormone therapy use and breast cancer incidence in non-Hispanic white women living in the Southwestern California from 2001 to 2004. J Clin Oncol, 25: 3437– United States. Breast Cancer Res Treat, 105: 327–335. 3439. doi:10.1200/JCO.2007.11.4132 PMID:17592152 doi:10.1007/s10549-006-9453-z PMID:17187234 Rosenberg LU, Einarsdóttir K, Friman EI et al. (2006b). Slattery ML, Sweeney C, Murtaugh M et al. (2005). Risk factors for hormone receptor-defined breast Associations between ERalpha, ERbeta, and AR geno- cancer in postmenopausal women. Cancer Epidemiol types and colon and rectal cancer. Cancer Epidemiol Biomarkers Prev, 15: 2482–2488. doi:10.1158/1055- Biomarkers Prev, 14: 2936–2942. doi:10.1158/1055- 9965.EPI-06-0489 PMID:17164374 9965.EPI-05-0514 PMID:16365013 Rosenberg LU, Magnusson C, Lindström E et al. (2006c). Slattery ML, Sweeney C, Wolff R et al. (2007b). Genetic Menopausal hormone therapy and other breast cancer variation in IGF1, IGFBP3, IRS1, IRS2 and risk of breast risk factors in relation to the risk of different histological cancer in women living in Southwestern United States. subtypes of breast cancer: a case-control study. Breast Breast Cancer Res Treat, 104: 197–209. doi:10.1007/ Cancer Res, 8: R11 doi:10.1186/bcr1378 PMID:16507159 s10549-006-9403-9 PMID:17051426 Rosenberg LU, Palmer JR, Wise LA, Adams-Campbell LL Stahlberg C, Pedersen AT, Andersen ZJ et al. (2004). Breast (2006a). A prospective study of female hormone use cancer with different prognostic characteristics devel- and breast cancer among black women. Arch Intern oping in Danish women using hormone replacement Med, 166: 760–765. doi:10.1001/archinte.166.7.760 therapy. Br J Cancer, 91: 644–650. PMID:15238982 PMID:16606813 Strom BL, Schinnar R, Weber AL et al. (2006). Case- Ross JA, Sinner PJ, Blair CK et al. (2005). Hormone control study of postmenopausal hormone replace- replacement therapy is not associated with an increased ment therapy and endometrial cancer. Am J Epidemiol, risk of leukemia (United States). Cancer Causes 164: 775–786. doi:10.1093/aje/kwj316 PMID:16997897 Control, 16: 483–488. doi:10.1007/s10552-004-7117-z Sweetman SC, editor (2008) Martindale: The Complete PMID:15986103 Drug Reference, London, Pharmaceutical Press, Rossing MA, Cushing-Haugen KL, Wicklund KG et al. Electronic version, (Edition 2008) (2007). Menopausal hormone therapy and risk of

281 IARC MONOGRAPHS – 100A

Thordarson G, Semaan S, Low C et al. (2004). Mammary Zahl PH & Maehlen J (2007). A decline in breast-cancer tumorigenesis in growth hormone deficient sponta- incidence. N Engl J Med, 357: 510–511, author reply 513. neous dwarf rats; effects of hormonal treatments. Breast PMID:17674462 Cancer Res Treat, 87: 277–290. doi:10.1007/s10549-004- 9504-2 PMID:15528971 Truong T, Orsi L, Dubourdieu D et al. (2005). Role of goiter and of menstrual and reproductive factors in thyroid cancer: a population-based case-control study in New Caledonia (South Pacific), a very high incidence area. Am J Epidemiol, 161: 1056–1065. doi:10.1093/aje/ kwi136 PMID:15901626 Tsukamoto R, Mikami T, Miki K et al. (2007). N-methyl- N-nitrosourea-induced mammary carcinogenesis is promoted by short-term treatment with estrogen and progesterone mimicking pregnancy in aged female Lewis rats. Oncol Rep, 18: 337–342. PMID:17611653 van Duijnhoven FJ, Peeters PH, Warren RM et al. (2006). Influence of and progesterone receptor polymorphisms on the effects of hormone therapy on mammographic density. Cancer Epidemiol Biomarkers Prev, 15: 462–467. doi:10.1158/1055-9965. EPI-05-0754 PMID:16537702 von Schoultz E & Rutqvist LEStockholm Breast Cancer Study Group. (2005). Menopausal hormone therapy after breast cancer: the Stockholm randomized trial. J Natl Cancer Inst, 97: 533–535. doi:10.1093/jnci/dji071 PMID:15812079 Waaseth M, Bakken K, Dumeaux V et al. (2008). Hormone replacement therapy use and plasma levels of sex hormones in the Norwegian Women and Cancer postgenome cohort - a cross-sectional analysis. BMC Womens Health, 8: 1 doi:10.1186/1472-6874-8-1 PMID:18194511 Weiss JM, Saltzman BS, Doherty JA et al. (2006). Risk factors for the incidence of endometrial cancer according to the aggressiveness of disease. Am J Epidemiol, 164: 56–62. doi:10.1093/aje/kwj152 PMID:16675538 Wigertz A, Lönn S, Mathiesen T et al.Swedish Interphone Study Group. (2006). Risk of tumors associated with exposure to exogenous female sex hormones. Am J Epidemiol, 164: 629–636. doi:10.1093/aje/kwj254 PMID:16835295 Wu AH, Yu MC, Tseng CC, Pike MC (2007). Body size, hormone therapy and risk of breast cancer in Asian-American women. Int J Cancer, 120: 844–852. doi:10.1002/ijc.22387 PMID:17131315 Yasui M, Laxmi YR, Ananthoju SR e t a l . (2006). Translesion synthesis past equine estrogen-derived 2′-deoxyadeno- sine DNA adducts by human DNA polymerases eta and kappa. Biochemistry, 45: 6187–6194. doi:10.1021/ bi0525324 PMID:16681391 Yuri T, Tsukamoto R, Uehara N et al. (2006). Effects of different durations of estrogen and progesterone treat- ment on development of N-methyl-N-nitrosourea- induced mammary carcinomas in female Lewis rats. In Vivo, 20: 6B829–836. PMID:17203775

282

Top View